Song Dissertation

Home , Acanthacris, Acherontiini, Aclerdidae, Aiolopus, Anacridium, Anacridium aegyptium

SYSTEMATICS OF CYRTACANTHACRIDINAE (ORTHOPTERA: ACRIDIDAE) WITH A FOCUS ON THE GENUS SCHISTOCERCA STÅL 1873: EVOLUTION OF LOCUST PHASE POLYPHENISM AND STUDY OF INSECT GENITALIA

DISSERTATION

Presented in Partial Fulfillment of the Requirements for

the Degree Doctor of Philosophy in the Graduate

School of The Ohio State University

Hojun Song, M.S.

*****

The Ohio State University 2006

Dissertation Committee: Approved by Dr. John W. Wenzel, Advisor

Dr. Norman F. Johnson ______

Dr. Johannes S. H. Klompen Advisor Graduate Program in Entomology

ABSTRACT

The systematics of Cyrtacanthacridinae (Orthoptera: Acrididae) is investigated to study the evolution of locust phase polyphenism, biogeography, and the evolution of male genitalia. In Chapter Two, I present a comprehensive taxonomic synopsis of the genus Schistocerca Stål. I review the taxonomic history, include an identification key to species, revise the species concepts of six species and describe a new species. In Chapter

Three, I present a morphological phylogeny of Schistocerca, focusing on the biogeography. The phylogeny places the desert locust S. gregaria deep within the New

World clade, suggesting that the desert locust originated from the New World. In Chapter

Four, I review the systematics of Cyrtacanthacridinae and present a phylogeny based on morphology. Evolution of taxonomically important characters is investigated using a character optimization analysis. The biogeography of the subfamily is also addressed. In

Chapter Five, I present a comprehensive review the recent advances in the study of locust phase polyphenism from various disciplines. The review reveals that locust phase polyphenism is a complex phenomenon consisting of numerous density-dependent phenotypically plastic traits. The evolution of locust phase polyphenism is investigated from a phylogenetic perspective. I show that different components of locust phase polyphenism evolved independently and often phylogenetically conserved. In Chapter

ii Six, I address some philosophical issues in the study of genital evolution. I show that the fundamental assumptions in the field of genital evolution are often violated and need to be tested more rigorously.

iii

To Vitaly Mikhailovich Dirsh (1904-1982),

An acridologist and a member of the phallic cult

What the chewing locust left, the swarming locust has eaten; What the swarming locust left, the crawling locust has eaten; And what the crawling locust left, the consuming locust has eaten (Joel 1:4, NKJV)

ACKNOWLEDGMENTS

My childhood dream was to be an entomologist. This dream has now become a

reality and I owe it all to my mentors, friends, colleagues, and family.

First and foremost, I would like to sincerely thank my advisor, Dr. John W.

Wenzel, for helping me to become a scientist. He is the best mentor I have ever met,

always willing to spare his time and always looking out for students’ best interest. After six years under his guidance, I sound a lot like him and I am proud of it!

I would also like to thank the members of my committee, Dr. Norman F. Johnson and Dr. Hans Klompen for providing guidance and support throughout my graduate career. In particular, I am grateful to Dr. Johnson for allowing me to use the collection and for discussions about taxonomy in general.

My understanding of grasshoppers and locusts improved greatly thanks to Dr.

Greg Sword. He welcomed me into his arms (in Sidney, Montana) when I was a budding graduate student and he has continued to encourage me throughout my career. Dr.

Theodore J. Cohn is one of the most generous and kind orthopterists I have ever known.

Whenever I needed help in pursuing my study, he was willing to help. I would like to express my gratitude for the late Dr. Reginald F. Chapman. He was the one who was

v truly interested in the evolution of Schistocerca and helped me shaping my ideas in many ways. He will be immensely missed.

I am grateful to my lab mates during my stay, Marc Branham, Sibyl Bucheli,

Ryan Caesar, Eric Dotseth, Todd Gilligan, Chi-Feng Lee, Kurt Pickett, and Joe

Raczkowski, and for putting up with my insanity. I am especially thankful to my office mate Sibyl for taking care of me like a brother. You made my stay ever more enjoyable. I also like to thank the members of the Museum of Biological Diversity for stimulating discussion and friendship.

My study would not have been complete without help of the curators the museums from where I borrowed specimens. Especially, I thank Dr. Dan Otte and Dr.

Jason Weintraub (ANSP) and Dr. Lacey Knowles and Mark O’Brien (UMMZ), Judith

Marshall (BMNH) for allowing me to work at their collections. Dr. John Capinera provided invaluable information on the ecology of grasshoppers during the fieldwork in

Florida. Department of the Army kindly allowed me to conduct a field study in

Oklahoma Fort Sill. Also, I would like to thank Dr. Zenón Cano-Santana (UNAM),

CONABIO, Secretaría de Marina, Armada de México for making a trip to Socorro Island possible. José Luis Castillo, Marcos Flores, Ivan Hernandez, and Enrique Arias provided invaluable assistance in the field. Dr. David B. Weissman and Dr. Ludividina Barrientos-

Lozano provided help in shaping the Socorro research into a manuscript.

My graduate study was supported by the National Science Foundation Graduate

Research Fellowship, OSU Alumni Grants for Graduate Research and Scholarship, OSU

Herta Camerer Gross Fellowship, and Tinker Field Research Grant.

vi On a personal note, I want to thank my kids at KUMC for helping me realize the meaning of life. I am grateful to my parents, Young-Gil Song and Jee-Yeon Koo, for giving me a chance and supporting me throughout my life. I want to shout out to my little brother Wonjun for encouraging me. My degree is really a fruit of my family’s prayers. I am indebted to my soul mate and best friend, Haeran Park, for believing in me.

After six years of studying grasshoppers and locusts, I feel like I only scratched the surface. It has been a humbling experience, but I remain hopeful because there are still many wonderful things out there for me to discover.

vii

VITA

March 31, 1977 …………………………… Born – Seoul, Korea

2000 ………………………………………. B.S. Entomology, Cornell University

2002 ………………………………………. M.S. Entomology, The Ohio State University

2002-2005 ………………………………… N.S.F. Graduate Research Fellow, The Ohio State University

PUBLICATIONS

Research Publication

1. Song, H., D. B. Weissman, L. Barrientos-Lozano, and Z. Cano-Santana. In Press. Locust Island. American Entomologist.

2. Song, H. In Press. Phylogenetic perspectives on the evolution of locust phase polyphenism. Journal of Orthoptera Research.

3. Song, H. 2006. Description of Schistocerca cohni n. sp. and redescription of S. socorro (Dirsh) (Orthoptera: Acrididae: Cyrtacanthacridinae) from Mexico. Zootaxa 1150: 43-52.

4. Weissman, D. B., H. Song, and L. Barrientos-Lozano. 2004. Locust Outbreak on Socorro Island, Islas Revillagigedo, México. TecnoINTELECTO 1: 119-121.

5. Song, H. 2004. On the origin of the desert locust Schistocerca gregaria (Forskål) (Orthoptera: Acrididae: Cyrtacanthacridinae). Proceedings of the Royal Society of London B. 271: 1641-1648.

6. Song, H. 2004. Revision of the Alutacea Group of genus Schistocerca (Orthoptera: Acrididae: Cyrtacanthacridinae). Annals of the Entomological Society of America 97: 420-436.

viii 7. Song, H. 2004. Post-adult emergence development of genitalic structures in Schistocerca Stål and Locusta L. (Orthoptera: Acrididae). Proceedings of the Entomological Society of Washington 106: 181-191.

8. Liebherr, J. K., and H. Song. 2002. Distinct ground beetle (Coleoptera: Carabidae) Assemblages within a New York State wetland complex. Journal of the New York Entomological Society 110: 127-141.

FIELDS OF STUDY

Major Field: Entomology

Specialization: Systematics of Acrididae, especially Cyrtacanthacridinae.

TABLE OF CONTENTS

Abstract ………….………………………………………………………………………. ii Dedication ………….…………………………………………………………………… iv Acknowledgments ………….………………………...…………………………………. v Vita ………….…………………..…………………………………………………….. viii List of Tables …………………..……………………………………..…………..……. xv List of Figures ………………..……………………………………..…………...……. xvii

Chapters:

1. Introduction ……………………………………………………………………… 1

2. Systematics of the locust genus Schistocerca Stål 1873 (Orthoptera: Acrididae: Cyrtacanthacridinae) …………………………………………………………….. 6

2.1. Introduction …………………………………………………………………. 6 2.2. Concept of the genus ……………………………………………………...… 7 2.3. Taxonomic history of the genus …………………………………………… 10 2.4. Biology and ecology of Schistocerca ……………………………………... 16 2.5. Variation of color ………………………………………………………….. 18 2.6. Identification key to Schistocerca species ………………………………… 21 2.7. A partial revision of the genus Schistocerca ………………………………. 37 2.7.1. Materials and methods ………………………………..………... 38 2.7.2. Taxonomic treatments ………………………………………..… 39 2.8. Taxon description ………………………………………………………….. 81

x 2.9. Conclusion ……………………………………………………………….. 147

3. Phylogeny, biogeography, and evolution of the locust genus Schistocerca Stål 1873 (Orthoptera: Acrididae: Cyrtacanthacridinae) ………………………….. 149

3.1. Introduction ………………………………………………………………. 149 3.2. Biogeography of Schistocerca and controversies ………………………... 150 3.3. Materials and methods ………………………………………………..….. 155 3.3.1. Taxon sampling ……………………………………………….. 155 3.3.2. Character sampling …………………………………………… 159 3.3.3. Phylogenetic analysis …………………………………………. 169 3.4. Results ……………………………………………………………………. 172 3.5. Discussion ……………………………………………………………...… 174 3.5.1. Generic relationships …………………………………………. 174 3.5.2. Species relationships ………………………………..………… 175 3.5.3. Discussion on major clades, synapomorphies and character evolution ……………………………………………………… 178 3.5.4. Phylogenetic interpretation of the biogeography of Schistocerca …………………………………………………... 189 3.6. Conclusion …………………………………………………..…………… 194

4. Systematics, character evolution, and biogeography of the bird-locust subfamily Cyrtacanthacridinae (Orthoptera: Acrididae) ……………………………….... 195

4.1. Introduction ………………………………………………………………. 195 4.2. Taxonomic review of Cyrtacanthacridinae ………………………………. 196 4.2.1. Subfamily Cyrtacanthacridinae ……………………………….. 196 4.2.2. Taxonomic discussion on cyrtacanthacridine genera ……….... 203 4.3. Materials and methods ………………………………………………...…. 223

xi 4.3.1. Taxon sampling ……………………………………………….. 223 4.3.2. Character sampling …………………………………………… 226 4.3.3. Phylogenetic analysis …………………………………………. 235 4.4. Results ……………………………………………………………………. 239 4.5. Discussion ………………………………………………………………... 239 4.5.1. Subfamily-level relationship within Acrididae ……………...... 239 4.5.2. Generic relationship within Cyrtacanthacridinae ……………... 242 4.5.3. Discussion on major clades, synapomorphies and character evolution ……………………………………………………… 245 4.5.4. Phylogenetic interpretation of the biogeography in Cyrtacanthacridinae ………………………………………...… 265 4.6. Conclusion ……………………………………………………………….. 269

5. A review of locust phase polyphenism and the phylogenetic perspectives on its evolution …………………………………………………………………….... 271

5.1. Introduction ………………………………………………………………. 271 5.2. History of phase theory …………………………………………………... 273 5.3. Recent advances in locust phase polyphenism research …………………. 280 5.3.1. Behavior ………………………………………………………. 280 5.3.2. Chemical ecology ……………………………………………... 292 5.3.3. Endocrinology ……………………………………………….... 307 5.3.4. Adaptive aspects of locust phase polyphenism ……………….. 315 5.4. Phylogenetic perspectives on the evolution of locust phase polyphenism ……………………………………………………………… 318 5.4.1. Current understanding ……………………………………….... 318 5.4.2. Review of locusts belonging to Cyrtacanthacridinae …………. 322 5.4.3. Phylogenetic perspectives on the evolution of locust phase polyphenism ………………………………………………...… 331

xii 5.5. Conclusion ……………………………………………………………….. 353

6. Test of fundamental assumptions in the study of genital evolution …………... 354

6.1. Introduction ………………………………………………………………. 354 6.2. Brief review of genital evolution ………………………………………… 355 6.2.1. Use of male genitalia in grasshopper taxonomy ……………… 355 6.2.2. Traditional hypotheses, new paradigm and controversies ……. 358 6.3. Systematics, genital evolution, and assumptions ………………………… 361 6.3.1. Genitalia are species-specific …………………………………. 362 6.3.2. Genitalia evolve at a rapid rate ……………………………….. 364 6.3.3. Genitalia are relatively invariable within a species …………... 367 6.4. Phylogenetic test of rapid genital evolution ……………………………… 369 6.4.1. Introduction …………………………………………………… 369 6.4.2. Phylogeny reveals the pattern of genital evolution ………….... 370 6.4.3. What is rapidity? ……………………………………………… 372 6.4.4. Test of rapid genital evolution based on literature data ………. 376 6.4.5. Composite nature of genitalia ………………………………… 380 6.4.6. Conclusion ……………………………………………………. 382 6.5. Geographic variation of male genitalia in bird-grasshopper Schistocerca lineata Scudder (Orthoptera: Acrididae: Cyrtacanthacridinae) …...……... 383 6.5.1. Introduction …………………………………………………… 383 6.5.2. Materials and methods ………………………………………... 386 6.5.3. Results ………………………………………………………… 393 6.5.4. Discussion …………………………………………………….. 400 6.6. Conclusion ……………………………………………………………….. 404

xiii Appendix A: ………………………………………………….……………………….. 406

Bibliography ……………………………………………………………..…………… 413

xiv

LIST OF TABLES

Table Page

2.1 Changes in taxonomic concepts in different revision of the genus. Partial revisions are not included in this table. Significant reduction in number between Kirby (1910) and Dirsh (1974) reflects a large number of synonymies. * indicates the species that may be valid, but not included in this work ………………………………………………………………... 9

2.2 A summary of taxonomic changes since Dirsh (1974). Harvey (1981) recognized five species and six subspecies that were considered to be subspecies of S. americana based on hybridization experiments. His synonymies are indicated in parenthesis. In this work, I revise one species and five species that Dirsh (1974) considered subspecies of S. americana and S. alutacea, respectively. I also describe a new species ……………….15

2.3 Four ecotypes of S. lineata and their color variation ……………………… 50

3.1 Taxon sampling included in the present phylogenetic analysis. Relevant taxonomic information, indicated by numbers, is presented below the table …………………………………………………………….. 157

3.2 A data matrix used to produce trees ………………………………………. 170

4.1 A brief summary of taxonomic changes within Cyrtacanthacridinae. I included all generic concepts attributed to the subfamily since 1870. In this work, I recognize 35 valid genera. Relevant taxonomic information, indicated by numbers, is presented below the table ………………………. 201

4.2 Taxon sampling included in the present phylogenetic analysis …………… 224

4.3 A data matrix used to produce trees ……………………………………….. 236

5.1 A summary of major findings in locust chemical ecology research since 1993 …………………………………………………………………. 293

xv 5.2 A list of acridid species commonly called locusts. They belong to at least six different subfamilies of Acrididae, indicating that locust phase polyphenism evolved multiple times. Information was taken from the International Society of Pest Information website (http://www.pestinfo.org/Literature/locspec.htm) ………………………… 321

5.3 Comparisons of phase related traits among the locust species in Cyrtacanthacridinae ……………………………………………………….. 324

5.4 A summary of density-dependent color change in known from Schistocerca species. Species in bold are locusts ………………………………………... 337

5.5 A summary of literature review on density-dependent color change known from Cyrtacanthacridinae. Normal behavior of some species is unknown, but likely to be sedentary, which is indicated by the asterisk ….. 344

6.1 A summary of the literature review. Of 89 publications we examined, 74 papers (80.9%) found genital characters useful in phylogenetic reconstruction. Number in parenthesis indicates percentage of papers that used genitalia for the particular order ……………………………………... 378

6.2 An analysis of genital character usage in the examined publications. Within percentage was calculated as a proportion of the number of publications within a certain category. Overall percentage was calculated as a proportion of the total number of publications (89 papers). a: Male genitalia used regardless of female genitalia use; b: female genitalia used regardless of male genitalia use ………………………………………………………….. 379

6.3 Allometric values and coefficient of variation found in the present study. r was calculated from Pearson correlation coefficient generated from a regression analysis of log-transformed data. The slope was calculated from the slope of a regression line of log-transformed data. Coefficient of variation was calculated as the standard deviation divided by the mean …. 396

xvi

LIST OF FIGURES

Figure Page

1.1 Two extreme phases of the Central American locust Schistocerca piceifrons (Walker). A typical gregarious phase is on the left and the solitarious phase is on the right …………………………...... 2

2.1 Two main morphological characters that distinguish Schistocerca from its relatives. A. bilobed male subgenital plate; B. quadrate male cerci ………. 8

2.2 A morphometric method Dirsh (1974) used in his revision. A pentagram was created using the proportional value of L, P, T, F, and C as percentage of the sum of these measurements. When such pentagrams representing species in question overlapped, he considered them to be conspecifics. Figure redrawn from Dirsh (1974: 40) ……………………………………..12

2.3 Key morphological characteristics of Schistocerca. Prosternal process is always present and mesosternal lobes are always elongated rectangular …. 22

2.4 Variation of antennae length in Schistocerca ……………………………... 24

2.5 Variation of tegmina length in Schistocerca ……………………………..... 25

2.6 Variation of prosternal process in Schistocerca …………………………....26

2.7 Variation of male cerci in Schistocerca ……………………………...... 26

2.8 Variation of hind margin of metazona of pronotum in Schistocerca ………28

2.9 Variation of male subgenital plate in Schistocerca ………………………... 29

2.10 Variation of male epiproct in Schistocerca ………………………………... 31

xvii 2.11 Schistocerca alutacea. A. lateral view of type; B. dorsal view of type; C. cingulum; D. endophallus; E. basal eminence of zygoma; F. epiphallus; G. male epiproct; H. male subgenital plate; I. male cercus; (Ac: arch of cingulum; Anc: ancora; Ap: apical valve of aedeagus; Apd: apodemes of cingulum; Bp: basal valve of aedeagus; Br: bridge of epiphallus; Cv: valve of cingulum; Gpr: gonopore process; Lp: lophus; Rm: rami of cingulum; Zyg: zygoma) ……………………………………………………………… 73

2.12 Schistocerca rubiginosa. A. lateral view of type; B. dorsal view of type; C. cingulum; D. endophallus; E. basal eminence of zygoma; F. epiphallus; G. male epiproct; H. male subgenital plate; I. male cercus ……………….. 74

2.13 Schistocerca lineata. A. lateral view of type; B. dorsal view of type; C. cingulum; D. endophallus; E. basal eminence of zygoma; F. epiphallus; G. male epiproct; H. male subgenital plate; I. male cercus ……………….. 75

2.14 Schistocerca shoshone. A. lateral view of type; B. dorsal view of type; C. cingulum; D. endophallus; E. basal eminence of zygoma; F. epiphallus; G. male epiproct; H. male subgenital plate; I. male cercus ……………….. 76

2.15 Schistocerca albolineata. A. lateral view of type; B. dorsal view of type; C. cingulum; D. endophallus; E. basal eminence of zygoma; F. epiphallus; G. male epiproct; H. male subgenital plate; I. male cercus ……………….. 77

2.16 Schistocerca obscura. A. lateral view of type; B. dorsal view of type; C. cingulum; D. endophallus; E. basal eminence of zygoma; F. epiphallus; G. male epiproct; H. male subgenital plate; I. male cercus ……………….. 78

2.17 Schistocerca cohni. A. lateral view of type; B. dorsal view of type; C. cingulum; D. endophallus; E. basal eminence of zygoma; F. epiphallus; G. male epiproct; H. male subgenital plate; I. male cercus ……………….. 79

2.18 Schistocerca socorro. A. lateral view of type; B. dorsal view of type; C. cingulum; D. endophallus; E. basal eminence of zygoma; F. epiphallus; G. male epiproct; H. male subgenital plate; I. male cercus ……………….. 80

xviii 3.1 Three different biogeographic hypotheses concerning the origin of the desert locust. In the map, green represents the distribution of the New World Schistocerca and red represents that of S. gregaria. Green terminals in the phylogeny represent the New World Schistocerca species while the red terminal with a square represents S. gregaria. A is a graphical representation of the New World Origin hypothesis where S. gregaria is nested within the New World Schistocerca. B represents the Old World Origin hypothesis where S. gregaria is basal to the New World species. C represents the Multiple Crossings hypothesis where the New World Schistocerca species are paraphyletic ………………….………………….. 152

3.2 A character coding scheme for prosternal process …….………………….. 162

3.3 A character coding scheme for male cercus ………………………………. 164

3.4 A character coding scheme for male subgenital plate. Character 29 was coded additively …………………………………………………………... 166

3.5 A strict consensus of four MPTs (L = 272, CI = 0.37, RI = 0.74). Monophyly of Schistocerca is shown in red. Numbers above the node are Bremer support values …………………………………………………….. 173

3.6 A strict consensus tree showing taxonomic instability of Dirsh’s (1974) revision. Many of his species concepts are shown to be paraphyletic in light of the present phylogeny …………………………………………….. 177

3.7 Major clades of interest in Schistocerca. Clades hypothesized to be results of a single colonization and radiation event are shown in color with appropriate geographic localities ………………………………………….. 179

3.8 A character optimization analysis for the color pattern on dorsal portion of pronotum. Colored dots with an arrow show the evolutionary trend in this character …………………………………………………………………… 183

3.9 A character optimization analysis for antennae length. The ancestral state for Schistocerca is the antennae as long as the combined length of head and pronotum. Longer antennae evolved repeatedly, and there was a drastic reduction in S. damnifica. This is DELTRAN optimization ……………… 186

3.10 A character optimization analysis for the shape of hind margin of pronotum. The ancestral state is the broadly round margin and angular margin evolved once, but its modification evolved multiple times ………………………… 187

xix 3.11 A character optimization analysis for the length of tegmina. The ancestral state is the tegmina extending about one pronotum length beyond abdomen. The lengthening and reduction of the tegmina evolved repeatedly. This is DELTRAN optimization …………………………………………………... 188

4.1 The shape of mesosternal lobes is the major synapomorphy for Cyrtacanthacridinae ……………………………………………………….. 197

4.2 Shape of frontal ridge of head …………………………………………….. 227

4.3 A character coding scheme for male epiphallus …………………………... 232

4.4 A character coding scheme for the shape of cingulum ……………………. 233

4.5 A strict consensus cladogram of 144 MPTs (L = 328, CI = 0.39, RI = 0.79). Monophyly of Cyrtacanthacridinae is shown in red. Black number above the node is Bremer support value. Gray number below the node is the number of unambiguous synapomorphies. Terminals with an asterisk are the taxa whose taxonomic concepts have changed in this work. (OED: Oedipodinae; GOM: Gomphocerinae; ACR: Acridinae; MEL: Melanoplinae; CAT: Catantopinae) ………………………………………………...... 238

4.6 A strict consensus phylogeny of Cyrtacanthacridinae showing major clades ………………………………………………………………………. 241

4.7 A character optimization analysis for the shape of frontal ridge of head. Frontal ridge elongated below the ocellus is the ancestral condition, and it is lost twice independently ………………………………………………. 250

4.8 A character optimization analysis of the shape of prosternal process. The ancestral condition is the cylindrical form, and L-shaped prosternal process evolved once ……………………………………………………… 251

4.9 The length of tegmina was reduced three times in Congoa, Halmenus, and Ootua. The arrow points to the reduced tegmina …………………….. 253

4.10 A character optimization analysis for the shape of male cerci. In many cases, the shape of male cerci is stereotypical and a good synapomorphy for a higher grouping ……………………………………………………… 255

4.11 Last abdominal segment of Cyrtacanthacridinae, showing epiproct, cerci, and male subgenital plate …………………………………………………. 256

4.12 Epiphalli of Cyrtacanthacridinae ……………………….…………………. 260 xx 4.13 Ectophallic sclerites of Cyrtacanthacridinae ……………………………… 261

4.14 Cingulum of Cyrtacanthacridinae ……………………………………….… 262

4.15 Endophalli of Cyrtacanthacridinae ………………………………………... 263

4.16 A reduced phylogeny of Cyrtacanthacridinae. This cladogram was generated by collapsing terminals down to genus. Next to each terminal is the main geographical distribution for each genus. Gray branch indicates ambiguous optimization. An asterisk next Schistocerca indicates that one species of the genus occurs in the Old World ……………………………... 266

5.1 Propensity to swarm is mapped as a binary present/absent character. Four locust species in Schistocerca do not form a monophyletic group. Character mapping analysis yielded two parsimonious scenarios of the evolution of swarming. This trait may have evolved three times and lost once (ACCTRAN) or evolved four independent times (DELTRAN) ………….. 334

5.2 Phylogeny of Cyrtacanthacridinae with a propensity to swarm optimized as a binary present/absent character. This trait appears to have evolved multiple times within this subfamily. Because Schistocerca clade is poorly resolved, it is difficult to determine how many times this trait has evolved, but see Fig. 5.1 …………………………………………………………………….. 340

5.3 Stepwise assembly of locust phase in Nomadacris-Patanga-Austracris clade. When locust phase is divided into smaller components and mapped onto the phylogeny, it is possible to hypothesize that many phase-related traits evolve differentially. Even traits that are commonly considered intimately associated such as behavior are dissociated in this clade ……………………………... 342

5.4 Evolution of color phenotypic plasticity in Cyrtacanthacridinae. When the propensity to swarm is mapped as a binary character, there are six independent origins, four of which are found in Schistocerca. Density dependent color change from literature data is mapped. Column A is the nymphal color in typical isolated settings. Column B indicates the change in background color (regardless of particular expressions) in response to change in density. Column C shows the development of black pattern in response to change in density ……………………………………………………………………... 350

5.5 Evolution of locust phase in Cyrtacanthacridinae. Sedentary, phenotypically plastic species represent an ancestral condition for the subfamily. From there, some lineages evolved behavioral plasticity to be locusts. Other lineages lost the plasticity all together …………………………………………………... 351

xxi 6.1 Phylogeny reveals the pattern of genital evolution. When male genitalia of eight Funkyphallus species (A) are mapped on to the phylogeny, a certain pattern emerges. Closely related species have more similar genital morphologies than more distantly related species (B). The morphological divergence between sister species can arise not only from rapid divergence, but from numerous extinction events (C) …………………………………. 371

6.2 A possible method to estimate the rate of genital evolution using phylogenetic reference and morphometric techniques ……………………. 375

6.3 Phallic complex of Schistocerca americana, modified from Song (2004b). (A) cingulum; (B) epiphallus; (C) ectophallic sclerite; (D) endophallus …. 381

6.4 Location of three allopatric populations of S. lineata used in the study ….. 387

6.5 Measurements used in the allometric study and the morphometric analysis. (A) length of pronotum to represent body size; (B) length of hind femur to study allometry; (C) width of apical valve of cingulum and width of basal eminence to represent genitalia size, and the image was used for morphometric analysis; (D) right cercus was used for morphometric analysis; (E) right lophus of epiphallus was used for morphometric analysis ………………………………………………….…………………. 392

6.6 Body size (A) and genitalia size (B) divergence among different populations. Red dots indicate mean value and asterisk indicates an outlier …………… 395

6.7 Geographic variation in male cerci. The scatter plot was generated by plotting PC 1 against PC2. Next to each axis, the contour of male cercus is shown. Black line is the mean value, and red and blue lines indicate two standard deviations. PC1 explained the width to length ratio the best, while PC2 explained the shape of bilobed the best. CO population (red) was significantly different from KS (green) and OK (blue) populations (ANOVA on PC2; F = 10.31; P < 0.0001) ………………………………... 397

6.8 Geographic variation in basal eminence of cingulum. The scatter plot was generated by plotting PC 1 against PC2. Next to each axis, the contour of male cercus is shown. Black line is the mean value, and red and blue lines indicate two standard deviations. PC1 explained the width to length ratio the best, while PC2 explained the shape of apical valves of cinglum the best. OK population (blue) was significantly different from CO (red) and KS (green) populations (ANOVA on PC2; F = 9.05; P < 0.0001) ………….… 398

xxii 6.9 Geographic variation in lophi of epiphallus. The scatter plot was generated by plotting PC 1 against PC2. Next to each axis, the contour of male cercus is shown. Black line is the mean value, and red and blue lines indicate two standard deviations. PC1 explained the width to length ratio the best, while PC2 explained the apical curvature of right lophus. CO population (red) was statistically different from KS (green) and OK (blue) populations (ANOVA on PC2; F = 3.79; P = 0.029) …………………………………………….... 399

xxiii

CHAPTER 1

INTRODUCTION

Locusts are the phylogenetically heterogeneous group of grasshoppers belonging to Acrididae (Orthoptera) that can form dense migrating swarms through extreme density-dependent phenotypic plasticity, in which cryptic, solitary individuals transform into conspicuous, gregarious individuals in response to change in population density (Fig.

1.1, Uvarov 1966, Pener 1983). In addition to color and behavior, morphology, endocrine action, biochemistry, nutritional intake, and genetic expression change in response to change in population density (Uvarov 1966, Pener 1991, Pener and Yerushalmi 1998,

Simpson et al. 1999). Locusts displaying the suite of characteristics associated with low and high local population densities are referred to be in solitarious and gregarious phases, respectively. This complex phenomenon of density-dependent phenotypic plasticity is known as locust phase polyphenism (Pener 1983).

One of the most well-known locusts is the desert locust Schistocerca gregaria. It

is the biblical locust recorded in Bible and Koran, and it is still affecting the lives of

many countries in Africa and the Middle East. In recent years, tremendous advances have

Figure 1.1: Two extreme phases of the Central American locust Schistocerca piceifrons (Walker). A typical gregarious phase is on the left and the solitarious phase is on the right.

been made in terms of understanding how behavior changes in response to density

(Simpson et al. 1999). Rigorous research programs are currently seeking to understand the chemical and physiological basis of locust phase polyphenism (Breuer et al. 2003,

Hassanali et al. 2005). Although the studies of the desert locust resulted in enormous knowledge about locust phase, our understanding of other locust species is still in its infancy. More than fifteen species of grasshoppers have been identified to express locust

2 phase polyphenism, but how locust phase evolved, whether different locust species

exhibit different phase characteristics, and how they are related remain unclear.

Strong flight capacity is one of the main features of locusts. This trait enabled

locusts to explore areas that are typically inaccessible to less mobile insects. Locusts have

been reported to fly across the oceanic barriers and over the high mountains. As a result,

many locust species often exhibit remarkable biogeographic patterns that can only be

explained by dispersal. For example, the genus Schistocerca contains about fifty species that are mainly distributed in the New World except for one species, S. gregaria, which occurs in the Old World. The red locust Nomadacris septemfasciata is confined to South

Africa and Madagascar, but all its relatives are known to be distributed in Australasia.

Both the evolution of locust phase polyphenism and the question of biogeography need to be understood from a historical perspective. Presently, locust phase polyphenism is considered to have evolved as an adaptation to heterogeneous environmental conditions brought on by high population density. However, several non-locust species are known to express phase-like traits, which is difficult to explain from an adaptive perspective alone. The biogeography of locust species has been addressed based on the anecdotal evidence and several hypotheses have been proposed, but they have not been tested explicitly. Thus, an explicit phylogenetic analysis is required to address these problems.

In this dissertation, I focus on the grasshopper subfamily Cyrtacanthacridinae, which contains several locust species. Especially, I study the genus Schistocerca in depth.

Schistocerca contains both locust and non-locust species and displays an intriguing

3 biogeographical pattern, which makes the genus an ideal group to study the evolution of locust phase polyphenism as well as the question of biogeography. A good phylogenetic analysis should be based on solid taxonomy. During the course of my dissertation, I discovered that the taxonomy of Schistocerca and Cyrtacanthacridinae is poorly understood. Therefore, I first establish sound taxonomy, which becomes the basis of the phylogenetic analysis. The resulting phylogeny provides a framework for understanding the evolution of locust phase and biogeography. The present phylogenetic analyses of

Schistocerca and Cyrtacanthacridinae are based on morphological characters. Male genital characters are often heavily used in the systematics of grasshoppers, and I use them in this analysis. In doing so, I became interested in the evolution of male genitalia, which resulted in a separate line of research.

Evolution of male genitalia is currently a very active area of research in evolutionary biology (Hosken and Stockley 2004). In most animals, the morphology of male genitalia is highly diverse and often species-specific, whereas that of female genitalia are relatively uniform. The current paradigm in understanding genital evolution is that genitalia are shaped by sexual selection (Eberhard 1985). Researchers of genital evolution make three implicit assumptions that are rarely tested: genitalia i) are species- specific; ii) evolve at a rapid rate; iii) are relatively invariable intra-specifically. During the course of my study, especially using Schistocerca, I found exceptions to all three assumptions and began to question their validity.

My dissertation addresses a range of topics from basic taxonomy, phylogeny, and evolutionary theories to some philosophical issues using grasshoppers as a study system.

4 It is organized in the following manner. Chapter Two and Three focus on the systematics

of the genus Schistocerca. Taxonomic review and species descriptions are included in

Chapter Two, and a phylogenetic analysis specifically addressing the biogeography of

Schistocerca is presented in Chapter Three. Chapter Four focuses on the systematics, biogeography, and character evolution of the subfamily Cyrtacanthacridinae. Using the phylogeny of Schistocerca and the phylogeny of Cyrtacanthacridinae, I address the evolution of locust phase polyphenism from a historical perspective in Chapter Five. A comprehensive review of recent advances in locust phase research is also included in this chapter. Finally, Chapter Six addresses the evolution of male genitalia. I review and test the fundamental assumptions of genital evolution and argue that it needs to be studied from a phylogenetic perspective.

CHAPTER 2

SYSTEMATICS OF THE LOCUST GENUS SCHISTOCERCA STÅL 1873

(ORTHOPTERA: ACRIDIDAE: CYRTACANTHACRIDINAE)

2.1 INTRODUCTION

Locust genus Schistocerca Stål (Orthoptera: Acrididae: Cyrtacanthacridinae) is a biologically fascinating group. It contains several locust species capable of forming an enormous swarm, consisting of billions of individuals, through density-dependent phase polyphenism (Uvarov 1966, Pener 1991). Most Schistocerca species are, however,

sedentary and non-swarming grasshoppers that are ecologically diverse and have adapted

to different environments. The genus is also known for its unusual biogeographic

distribution where a single species, S. gregaria (Forskål) occurs in the Old World, while

the rest of the genus occurs strictly in the New World. This transatlantic disjunction

sparked heated debates concerning the origin of the desert locust among scientists (Kevan

1989, Ritchie and Pedgley 1989, Song 2004b, Lovejoy et al. 2006).

Despite attractive aspects of the genus, its taxonomy is not well understood. This

dichotomy can certainly be attributed to the variability within a species and the similarity

6 across the species. At one extreme, some species are widespread and highly variable in terms of color and ecology (Hubbell 1960). At another extreme, certain closely related species are externally and internally so similar that they were synonymized to be conspecifics (Dirsh 1974). The lack of taxonomic resolution hinders the study of

Schistocerca and here I attempt to improve the systematics of the genus.

In this chapter, I present the systematic overview of the genus Schistocerca. The taxonomic history of the genus is reviewed. Biology, ecology, and the variation of color are discussed. I also present a comprehensive taxonomic identification key of the genus for the first time. Finally, I revise several taxonomic concepts and describe a new species based on examinations of museum specimens and field observations.

2.2 CONCEPT OF THE GENUS

The genus Schistocerca Stål 1873 belongs to a grasshopper subfamily

Cyrtacanthacridinae (Orthoptera: Acrididae). The subfamily is characterized by the rectangular inner margins and acute inner angles of the mesosternal lobes, which is well- expressed in Schistocerca (Uvarov 1923). Schistocerca is one of two New World representatives of the largely Old World subfamily. Another New World genus is

Halmenus Scudder 1893, which has been suggested to be the closest relative of

Schistocerca (Dirsh 1974).

Schistocerca is the largest genus of Cyrtacanthacridinae containing about 50 species. Forty-six species are recognized and treated in the present study (Table 2.1). The

7 main morphological character that distinguishes Schistocerca from the rest of the subfamily is the shape of male subgenital plate (Fig. 2.1A). The genus has a bilobed male subgenital plate whose lobes are distinctly protruding. The shape of male cerci is quadrate in its basic form although there are many modifications (Fig. 2.1B).

Figure 2.1: Two main morphological characters that distinguish Schistocerca from its relatives. A. bilobed male subgenital plate; B. quadrate male cerci.

Scudder, 1899 Kirby, 1910 Dirsh, 1974 This work (43 concepts) (72 concepts) (40 concepts) (49 concepts)

aequalis obliquata aequalis impleta subvittata alutacea albolineata matogrosso albolineata nitens (Guatemala) albolineata obscura albolineata infumata tatarica alutacea alutacea melanocera alutacea nitens (Nuevo Leon) alutacea pallens alutacea inscripta tolteca alutacea lineata nitens carribeana americana nitens (Trinidad) americana paranensis ambigua interrita vaga alutacea rubiginosa nitens colombina beckeri nitens (Vera Cruz) aurantia peregrina americana lineata varipes alutacea shoshone nitens nitens braziliensis obscura australis pyramidata aurantia literosa venusta alutaceca insignis nitens virginica brevis* orinoco* bivittata rubiginosa australis luridescens vicaria americana americana obscura camerata pallens (Barbados) bogotensis separata bivittata maculipennis viridescens americana benedicto orinoco cancellata pallens (Bolivia) cameraa shoshone bogotensis malachitica viridis americana cancellata pallens carribeana pallens (Ecuador) cancellata simulatrix brachyptera maya vittagrons americana cubense quisqueya centralis pallens (Mexico) carinata sonorensis camerata melanocera vittata americana flaviventris subspurcata ceratiola piceifrons peruviana columbina vaga cancellata mexicana vitticeps americana gregaria cohni piceifrons piceifrons crocotaria venusta carinata milberti vittigera americana interrita columbina quisqueya damnifica zapoteca carneipes obliquata zapoteca americana paranensis crocotaria rubiginosa

9 desiliens columbina obscura americana peruviana cubense serialis exsul cribrata olivacea americana serialis damnifica shoshone flavofasciata cristagalli pallens americana socoro diversipes socorro gracilis crocotaria paranensis beckeri flavofasciata subspurcata gulosa cubensis patiana braziliensis gorgona vaga idonea damnifica perturbans brevis gregaria flaviventris virginica infumata democratica peruviana camerata gregaria gregaria inscripta desiliens philippina centralis interrita interrita emortualils rustica ceratiola lineata lineata exsul semivittata damnifica literosa literosa flavofasciata separata diversipes magnifica* maya frontalis septentrionalis flavofasciata matogrosso* melanocera gracilis shoshone gorgona melanocera mellea gulosa sonorensis literosa nitens (Colima) mexicana idonea subspeurcata magnifica nitens (Colombia)

Table 2.1: Changes in taxonomic concepts in different revision of the genus. Partial revisions are not included in this table. Significant reduction in number between Kirby (1910) and Dirsh (1974) reflects a large number of synonymies. * indicates the species that may be valid, but not included in this work.

2.3 TAXONOMIC HISTORY OF THE GENUS

The genus Schistocerca has been partially or completely revised several times,

and overall it has a convoluted taxonomic history. A brief review of taxonomic history

from 1873 to 1974 can be found in Dirsh’s revision (Dirsh 1974: 31) and here I briefly

summarize it and add recent findings.

The genus Schistocerca was described by Stål in 1873 as a subgenus of the genus

Acridium. Dirsh (1974) suggested that Stål raised its rank to a genus in 1876, but Vickery and Kevan (1983) suggested that Stål still considered it as a subgenus of Acridium. It was

fully recognized as a genus by Brunner von Wattenwyl in 1882. Scudder (1899) was the

first to revise Schistocerca, with 44 species. He relied mostly on color pattern to describe

species. Kirby (1910) in his synonymic catalogue of Orthoptera listed 73 species of

Schistocerca, which reflected numerous new descriptions between 1899 and 1910. Since

then, numerous regional studies were published which treated the genus partially (United

States: Rehn 1902a, Rehn and Hebard 1909a, b, Hubbell and Walker 1928, Hebard 1932a,

1935, Henderson 1942, Tinkham 1948, Hubbell 1960; Galápagos Islands: Snodgrass

1902, Dirsh 1969; Costa Rica: Rehn 1905; Paraguay: Bruner 1906; South America:

Bruner 1911, 1913, 1920; Argentina: Rehn 1913a; Colombia: Hebard 1923, 1933a;

Panama: Hebard 1924, 1933b; Mexico: Rehn 1913b, Hebard 1925, 1932b; West Indies:

Rehn and Hebard 1938; Canada: Vickery and Kevan 1964). In these studies, some

synonymies were made, again mostly based on color pattern.

In 1960, Hubbell published a detailed revision on the sibling-species of the

Alutacea Group. He found that male phallic structures were particularly useful in

distinguishing species. Although he briefly discussed the genus as a whole, Hubbell was

mostly interested in North American species. He recognized seven species groups of

Schistocerca occurring in North America north Mexico: Ceratiola, Damnifica, Americana,

Vaga, Alutacea, Shoshone, and Obscura Groups. Hubbell considered Ceratiola and

Damnifica Groups to be distinct on their own, with no close relatives. He suggested that

Americana and Vaga groups were closely related on the basis of having short antennae, a

supra-anal plate with median sulcus not abruptly terminated by a tranverse carina and

without distinct admesal prominences, and a V-notched male subgenital plate. He

regarded the remaining three species groups to be a monophyletic group on the basis of

unusually long antennae. Hubbell also demonstrated that color pattern is highly variable

and often influenced by local environments.

Dirsh (1965) published a preliminary note for the revision in which he studied

numerous types in which he proposed 27 new synonyms. This work was based on direct

comparisons of type materials and numerous museum specimens. Many of the taxonomic

concepts he synonymized were of Francis Walker.

In 1974, Dirsh published a comprehensive revision of Schistocerca as a book, with included 22 species and 21 subspecies. In this revision, he based species descriptions on a morphometric method modified from a technique that was used to distinguish between two phases of a locust (Dirsh 1953). The method works as follows: lengths of body (L), pronotum (P), tegmen (T), and hind femur (F), and widths of hind

femur (Fw) and head (C) are measured for each species and the mean value for each dimension is calculated. For each species, the proportional value of L, P, T, F, and C is calculated as a percentage of the sum (i.e. L / (L+P+T+F+C) for each species). These proportional values are arranged around a circle so that the circle is equally divided in

Figure 2.2: A morphometric method Dirsh (1974) used in his revision. A pentagram was created using the proportional value of L, P, T, F, and C as percentage of the sum of these measurements. When such pentagrams representing species in question overlapped, he considered them to be conspecifics. Figure redrawn from Dirsh (1974: 40).

five pie-shaped pieces (72 internal degrees each) and the five lines dividing the circle are

scaled with the center of the circle being 0%. By plotting and connecting the proportional

values on the lines for each species, a pentagram is drawn, which becomes a graphical

representation of a morphometric species (Fig. 2.2). If these pentagrams are similar and

overlap, they are considered to be the same species. No statistic analysis is associated

with this method. In the end, Dirsh (1974) synonymized 11 names under S. americana,

16 names under S. nitens, and six names under S. alutacea. For S. americana and S. aluacea, he retained the previous taxonomic concepts and lowered the taxonomic ranks to subspecies. For S. nitens, he only recognized some island forms, but did not recognize the rest. Together with four names Dirsh synonymized in his earlier work (Dirsh 1965),

20 taxonomic concepts were synonymized under S. nitens based on the morphometric

method. Dirsh’s (1974) work is the most comprehensive revision on Schistocerca

taxonomy to date, containing important information on type specimens and all the

relevant literature along with detailed drawings of male and female genitalia for each

species.

Dirsh’s (1974) revision considered all the agriculturally important locust species

(S. gregaria, S. piceifrons, and S. cancellata) to be the subspecies of an American species

S. americana. This taxonomic action had major practical implications because numerous

studies had already been published on locust species such as S. gregaria. It also had

interesting evolutionary implications because the desert locust from Africa, S. gregaria

was considered to be a subspecies of a non-swarming American species. In order to test

Dirsh’s species concept, several scientists at the Centre for Overseas Pest Research

(COPR) carried out hybridization experiments using several subspecies of S. americana sensu Dirsh (1974). Harvey (1979) hybridized locusts from the Yucatan peninsula of

Mexico with locusts from Argentina and Florida. He found that the sex ratios, fertility, and meiosis in F1 hybrids were severely affected and concluded that the locusts from three geographic localities represented three biologically distinct species. Jago et al.

(1979) hybridized three New World Schistocerca species (what are now known as S. americana, S. cancellata, and S. pallens) with the African S. gregaria. They found that in most cases the hybridization experiments produced over 90% non-viable eggs with no developing embryos. The cross between S. pallens females and S. gregaria males produced about 86% non-viable eggs while the cross between S. americana females and

S. gregaria males produced about 48% non-viable eggs. Based on these results, Jago et al.

(1979) argued that S. gregaria was a biologically distinct species and could not be considered a subspecies of S. americana. Harvey (1982) and Jago et al. (1982) published further results from their hybridization experiments separating S. piceifrons and its subspecies from S. cancellata.

Harvey (1981) reclassified the species that Dirsh (1974) considered to be the subspecies of S. americana based on the hybridization experiment results at COPR and the study of museum specimens. In this work, he recognized six species to comprise the

Americana Complex and they were S. americana, S. serialis (with two subspecies), S. piceifrons (with two subspecies), S. cancellata, S. gregaria (with two subspecies), and S. pallens. Although Dirsh (1974) considered S. pallens to be a distinct species from S. americana, Harvey (1981) included it in his reclassification because it could readily

hybridize with S. cancellata, indicating a close relationship. Harvey (1981) excluded

three nominal species which Dirsh (1974) treated as subspecies of S. americana. They

were S. interrita, S. maculata (which Dirsh (1974) synonymized with S. interrita), and S. americana socorro.

Dirsh, 1974 This work alutacea (6 ssp.) alutacea, stat. rev. americana (11 ssp.) rubiginosa, stat. rev. beckeri lineata, stat. rev. braziliensis shoshone, stat. rev. brevis albolineata, stat. rev. camerata (= insignis, syn.) centralis cohni n. sp. ceratiola socorro, n. stat., just. emend. damnifica diversipes flavofasciata Harvey, 1981 gorgona americana literosa cancellata magnifica (= paranensis) matogrosso gregaria gregaria melanocera gregaria flaviventris nitens (4 ssp.) piceifrons piceifrons obscura (= americana benedicto) orinoco piceifrons peruviana pallens serialis serialis quisqueya serialis cubense subspurcata

Table 2.2: A summary of taxonomic changes since Dirsh (1974). Harvey (1981) recognized five species and six subspecies that were considered to be subspecies of S. americana based on hybridization experiments. His synonymies are indicated in parenthesis. In this work, I revise one species and five species that Dirsh (1974) considered subspecies of S. americana and S. alutacea, respectively. I also describe a new species.

The hybridization studies collectively demonstrated that many concepts presented

in Dirsh’s revision are taxonomically unstable (Table 2.2). Dirsh (1974) included six

subspecies of S. alutacea and 16 taxonomic concepts synonymized under S. nitens. I test

Dirsh’s concept on S. alutacea in this chapter based on examination of discrete

morphological characters. This work was published as a form of a partial revision (Song

2004a). No similar study is available to test Dirsh’s concept on S. nitens. Therefore, a

comprehensive revision of the genus is much needed.

2.4 BIOLOGY AND ECOLOGY OF SCHISTOCERCA

The genus Schistocerca is biologically interesting within Acrididae in that it

includes both locusts and sedentary grasshoppers. An enormous amount of ecological

information is available on the locust species, especially of the desert locust, S. gregaria.

A detailed review of biology of locust species is presented in Chapter 5. In this section, I

briefly review the biology and ecology of sedentary Schistocerca species.

Schistocerca belongs to the grasshopper subfamily Cyrtacanthacridinae, which is generally known to favor arboreal habitat and prefer to feed on herbaceous plants rather than grassy plants (Dirsh 1974, Uvarov 1977). Sedentary Schistocerca species exhibit the similar trend, but moreover, there are indications that the host plant association plays a major role in the evolution of Schistocerca (Sword 1999, Sword and Dopman 1999,

Sword et al. 2000, Dopman et al. 2002).

Within Schistocerca, a spectrum of host plant association, from strict monophagy

to broad polyphagy, has been reported. At an extreme end is S. ceratiola, a nocturnal and

monophagous species endemic to central Florida (Hubbell and Walker 1928). Both field

observations and feeding experiments confirmed that it is strictly monophagous on

Florida rosemary, Ceratiola ericoides (Smith and Capinera 2005). At another end of the spectrum is the desert locust S. gregaria. The swarms of the desert locust have been

reported to feed on some 400 species of plants (Uvarov 1977). However, more explicit

feeding experiments suggest that S. gregaria do indeed exhibit some preference for

certain plants. Other locust species in the genus display a similar pattern of feeding to the

desert locust. Sedentary species in the genus, however, exhibit rather interesting feeding

habits.

It has been known that sedentary Schistocerca species are narrowly oligophagous.

Criddle (1932) reported that a Canadian population of S. lineata showed a marked

preference for American licorice Glycyrrhiza lepidota, and readily accepted Astragulus,

Lathyrus and Vicia. Kevan (1943) found that S. flavofasciata in Trinidad fed on fifteen

species of plants, but it showed a preference for maize and banana leaves. Duck (1944)

performed a feeding experiment on S. obscura using 22 species of plants. He found that

the species had a marked preference toward American elm (Ulmus americanus) and

Cotton (Gossypium herbaceum). He noted that nymphs fed on the low scrubby elm

growth until maturity, suggesting a host-plant association. Sword and Chapman (1994)

demonstrated that there were two distinct host races in S. shoshone, one on jojoba

(Simmondsia chinensis) and another on mesquite (Prosopis velutina). Sword and

Dopman (1999) studied two host races of S. lineata in Texas that were highly associated with Ptelea trifoliata and Rubus trivialis, respectively. I found a population of S. lineata from Colorado that almost exclusively feeds on Tamarisk (Tamarix sp.), and another population in Kansas that seemed to be associated with poison sumac (Rhus vernix)

(Song, unpublished data).

Nymphal and adult instars of the hemimetabolous insects are generally considered to have the similar feeding habit. Studies on Schistocerca challenge this idea. Sword and

Dopman (1999) found that Texas populations of S. lineata showed ontogenetic specialization on certain host plants. They demonstrated that the nymphs of polyphagous

S. lineata were in fact monophagous and this pattern is associated with a geographic structure (Dopman et al. 2002). Nymphal specialization on toxic plants might be adaptive when predation rate is high. However, grasshoppers also require balanced nutrients for improved growth, thus adults would shift to broader spectrum of food plants (Sword and

Dopman 1999). Mounting data suggest that the ontogenetic specialization is rather widespread in Schistocerca (Sword and Chapman 1994, Sword 1999, Sword and

Dopman 1999, Sword et al. 2000, Smith and Capinera 2005).

2.5 VARIATION OF COLOR

Schistocerca is known as a taxonomically difficult group mainly because of its color variation (Hebard 1923, Hubbell 1960). Variability of color in grasshoppers is well known and sometimes ineffective in differentiating species (Hubbell 1960) and this is

especially true for the North American Schistocerca species. Hubbell (1960) analyzed the variability of color patterns in S. lineata, S. alutacea, and S. rubiginosa (Harris), and here

I briefly review his work and other studies (King and Slifer 1955, Chapman et al. 1995,

Sword and Dopman 1999).

The dorsal stripe, which extends from the vertex of the head to the anterior tip of

pronotum, is one of several variable color patterns in Schistocerca (Rehn 1901, 1902b).

In early 1900, S. alutacea was known to be always striped and S. rubiginosa was thought to be unstriped. Rehn (1901, 1902b) observed a striped form and an unstriped form in copula and suspected that S. rubiginosa was just an unstriped form of S. alutacea. After examining a large number of specimens of both species identified by unique phallic morphologies, Hubbell (1960) concluded that S. alutacea is always striped and S. rubiginosa can be either striped or unstriped. He then plotted the occurrence of a dorsal stripe in S. rubiginosa against the collecting localities of the specimens and found that the southern populations gradually lose the dorsal stripe. Similar phenomenon is found in S. shoshone (Thomas). Originally, the species was described based on the striped form and

Scudder (1899) described two additional species based on the absence of the dorsal stripe.

Hebard (1935), when synonymizing unstriped S. venusta Scudder under striped S. shoshone, hypothesized that S. venusta was nothing more than an unstriped form of S. shoshone. Chapman et al. (1995) suggested that these two forms are associated with different host plants. Individuals that feed on jojoba, Simmondsia chinensis, were always striped and individuals that feed on mesquite, Prosopis velutina, were always unstriped.

Chapman and Sword (unpublished data) performed several crossing experiments between

striped and unstriped forms, and found that offspring were always striped regardless of the sex of the parents, confirming Hebard’s (1935) hypothesis on the variability of the stripe.

Hubbell (1960) also studied the regional variations in body color, markings of head and thorax, mesepimeral stripe, tegminal maculation, and hind tibiae. He demonstrated that the different populations of a single species could vary considerably in color patterns and that color alone could not be used as a definite diagnostic character.

For example, he showed that the color of hind tibiae varied according to the populations, especially in S. lineata. The color of the hind tibiae was yellowish or brownish in the eastern populations, coral pink or red in the midwestern populations, and black in the southwestern populations. A similar phenomenon can be observed in S. albolineata

(Thomas), whose Arizona populations have red hind tibiae, while the Texas and Mexican populations have black hind tibiae (Song, unpublished data).

Schistocerca lineata is one of the most polymorphic species in the genus and

Hubbell (1960) showed that the coloration intensified toward southwestern regions.

Populations in Illinois, Indiana, Michigan, and Ohio were lightly colored, whereas those in Oklahoma and Texas were conspicuously colored. He hypothesized that the variation could be associated with the food preference of a local population, and not necessarily related to the habitat. Dirsh proposed that the conspicuously colored form was a hybrid between S. albolineata and S. lineata (Dirsh 1974: 191, 211), which he used as a reason for his synonymy. However, this Texas population is not a hybrid, but an aposematic population of S. lineata that probably feeds on toxic Ptelea trifoliata L. (Rutaceae)

(Sword and Dopman 1999). Sword and Dopman (1999) studied several populations of S. lineata in Texas that utilized different host plants as a nymphal instar. They showed that the intensity of density-dependent aposematism was different in populations associated with toxic P. trifoliata and populations associated with palatable Rubus trivialis Michaux

(Rosaceae). When crowded, nymphs from both populations produced an aposematic coloration, but Ptelea-associated nymphs were more intense in color. Sword (2002) suggested that the nymphal coloration associated with a host plant is genetic, not necessarily related to what nymphs feed on, because he was able to rear different color phenotypes on the same standard diet (lettuce). The ontogenetic host specialization in relation to the aposematism is only known in Texas populations of S. lineata, and, given the species’ wide distribution, midwestern and eastern populations might have different ecologies (Sword and Dopman 1999).

2.6 IDENTIFICATION KEY TO SCHISTOCERCA SPECIES

Below I present an identification key to 43 Schistocerca species. In creating this key, I often relied on phylogenetically useful, but diagnosable external characters, although there were some genital characters I included. Illustrations of useful morphological characters are also included. Not included are S. brevis, S. matogrosso, S. orinoco and S. magnifica, which I did not have access to. Known distribution of the species is indicated in brackets.

Figure 2.3: Key morphological characteristics of Schistocerca. Prosternal process is always present and mesosternal lobes are always elongated rectangular.

1. Mesosternum with the length of lateral lobes longer than their width (Fig. 2.3).

Prosternal process always present (Fig. 2.3). Male subgenital plate always bilobed

(Fig. 2.1). Usually moderate to large size [Entirely in New World except S.

gregaria] ...…...... ……………………………………..….. 2, Schistocerca Stål

Mesosternum with the length of lateral lobes as wide as or wider than their width.

Prosternal process present or absent. Male subgenital plate variable. Usually small to

medium size …………………………………………...….…...... … Other Acrididae

2. Antennae length, especially in males, shorter or as long as the combined length of

head and pronotum (Fig. 2.4A, B)….…………...…………………………………… 3

Antennae length, especially in males, visibly longer than the combined length of head

and pronotum (Fig. 2.4C, D) .…………………...………………………………….. 19

3. Antennae shorter than the combined length of head and pronotum and weakly

ensiform (Fig. 2.4A). Median carina of pronotum distinctly elevated. Tegmina,

especially in females, as long as or slightly shorter than the tip of abdomen. Entire

body rusty brown sometimes with slight mottling on tegmina. [Southeastern U.S. and

northern Mexico] ……….……………….………...……...…. S. damnifica (Saussure)

Antennae as long as the combined length of head and pronotum (Fig. 2.4B) and

median carina of pronotum not distinctly elevated ………………..………..……….. 4

4. Length of tegmina only slightly extending beyond abdomen (Fig. 2.5A) ………...… 5

Length of tegmina visibly extending beyond abdomen (Fig. 2.5B, C) …………….... 7

5. Shape of basal eminence of cingulum rounded rectangular. Median carina low.

Overall color light brown. Tegmina with faint maculation. Outer surface of hind

femora without dark bands. Similar to the Americana Complex, but much smaller in

size. [Entire Central America] …………………………...... ……… S. centralis Dirsh

Shape of basal eminence of cingulum round. Median carina distinctly raised as a

ridge. Found in West Indies ……….…………………………..………………...... … 6

Figure 2.4: Variation of antennae length in Schistocerca. 24

Figure 2.5: Variation of tegmina length in Schistocerca.

6. Prosternal process distinctly curved backward, nearly touching thoracic sternum (Fig.

2.6B). Apex of male cerci distinctly bilobed (Fig. 2.7D). A large black area on the

inner surface of the hind femora present. [Cuba, Jamaica, Bahamas] ……………...…..

…………………………………………………………………. S. cubense (Saussure) 25

Prosternal process straight or only slightly curved backward, never touching thoracic

sternum (Fig. 2.6A). Apex of male cerci only lightly bilobed (Fig. 2.7C). Inner

surface of hind femora without distinct black markings. [Hispaniola, Puerto Rico and

islands east to Antigua] ……………………………………..… S. serialis (Thunberg)

Figure 2.6: Variation of prosternal process in Schistocerca.

Figure 2.7: Variation of male cerci in Schistocerca.

7. Length of tegmina extending distinctly more than the length of one pronotum beyond

abdomen (Fig. 2.5C). Tegmina transparent with large distinct dark patches ……...... 8

Length of tegmina extending only about the length of one pronotum beyond abdomen

(Fig. 2.5B). Tegmina pattern variable .……….…………………………………..… 12

8. Lophi of epiphallus simple right angular in shape ………...………………..……….. 9

Lophi of epiphallus right angular in shape with inner side inflated ………...……... 11

9. Apical valve of cingulum of endophallus slightly protruding from cingulum. Median

carina low. Central American locust. [Mexico south of the Tropic of Cancer and

Central America, with a subspecies S. piceifrons peruviana in Peru and southern

Ecuador] ………………………………………………………. S. piceifrons (Walker)

Apical valve of cingulum of endophallus distinctly protruding from cingulum.

Median carina raised as a ridge ……………………………………………….....…. 10

10. Pronotum with dark brown lateral stripes with a creamy yellow dorsal stripe in the

middle running from head to tegmina. Hind tibiae usually orange red. [Southern U.S.

from Texas to Florida] ………………………………………… S. americana (Drury)

Median carina of pronotum with deep sulci. Hind tibiae brownish. Only found in the

Old World. Desert locust. [Northern Africa and Middle East, with a subspecies, S.

gregaria flaviventris found in the Republic of South Africa] ….. S. gregaria (Forskål)

11. Upper carinula of hind femora with prominently protruding granules. Prosternal

process straight or only slightly curved backward, never touching thoracic sternum

(Fig. 2.6A). Male cerci distinctly bilobed (Fig. 2.7D). Male subgenital plate incision

shallow. Apical valve of cingulum simply narrowing toward apex. South American

locust. [Chile, Argentina, Bolivia, Paraguay, Uruguay and southern

Brazil] ……………………………………………………….... S. cancellata (Serville)

Upper carinula of hind femora with miniscule granules. Prosternal process distinctly

curved backward, nearly touching thoracic sternum (Fig. 2.6B). Male cerci only

lightly bilobed (Fig. 2.7C). Male subgenital plate incision deep. Apical valve of

cingulum strongly curved up. [Bolivia] ...………... S. pallens (Thunberg) BOLIVIA

Figure 2.8: Variation of hind margin of metazona of pronotum in Schistocerca.

12. Posterior margin of metazona of pronotum broadly round (Fig. 2.8A) ………….… 13

Posterior margin of metazona of pronotum angular (Fig. 2.8B, C) ………………..…... 17

13. Male subgenital plate incision U-shaped (Fig. 2.9A) ……………………………… 14

Male subgenital plate incision V-shaped (Fig. 2.9B) ………………………..………… 15

Figure 2.9: Variation of male subgenital plate in Schistocerca.

14. Frontal half of pronotum dark green and distal half bright yellow, with bright yellow

dorsal stripe. Base of tegmina with bright orange veins. Hind femora overall yellow

with orange and dark green markings. Tegmina often smoked black. [Endemic to

Galápagos Islands, Ecuador; on all islands except Española] .….. S. melanocera (Stål)

Pronotum brownish without dorsal stripe. Hind femora brownish with two dark bands.

Tegmina semi-transparent with brown maculation. [Endemic to Galápagos Islands,

Ecuador; Islas Española, Floreana, Gardner near Española, Genovesa, San

Cristóbal] ………………………………………………...….....… S. literosa (Walker)

15. Integument sculpting pattern on pronotum with thickened ridges and shallow

punctures. Pronotum with granules. Tegmina with faded maculation. Hind femora

with two dark bands. Hind tibae red. [Trinidad] ……………………. S. beckeri Dirsh

Integument sculpting pattern on pronotum fading near median carina ……….....… 16

16. Pronotum with granules present on dorsal surface. Prosternal process distinctly

curved backward, nearly touching thoracic sternum (Fig. 2.6B). Tegmina with

elongate, faded maculation. Apical valve of cingulum simply narrowing toward apex.

[Mexico] ………………………………………….. S. pallens (Thunberg) MEXICO

Pronotum without granules. Prosternal process straight or only slightly curved

backward, never touching thoracic sternum (Fig. 2.6A). Tegmina with large brown

maculation. Apical valve of cingulum rectangular and distinctly curved down. Male

subgenital plate notch deep and angular. [Ecuador and Peru] ………………………….

………………………………………………………………. S. subspurcata (Walker)

17. Posterior margin of metazona of pronotum acutely angular with pointed apex (Fig.

2.8C). Overall color brown. Lateral lobe without distinct pattern. [Endemic to

Socorro Island, Islas de Revillagigedo, Mexico] ...... … S. socorro (Dirsh)

Posterior margin of metazona of pronotum obtusely angular with round apex (Fig.

2.8B). Color variable ……...……………………………………….……………….. 18

18. Tegmina with mottled maculation. Overall gray in color. Lateral lobe of pronotum,

with single dark marking and white marking at the base. Hind femur with two dorsal

bands. [Western U.S., including California, Arizona, New Mexico, Texas, Kansas,

Oklahoma, and the northern Mexico] ……...………………………. S. vaga (Scudder)

Tegmina with large pantherine maculation. Overall brownish in color. Yellow dorsal

stripe running from head to pronotum, disappearing in the middle of pronotum.

[Peru] ………………………………………………………....… S. interrita Scudder

19. Antennae length, especially in males, extending much beyond pronotum (Fig.

2.4D) ……………………………………………………………………………….. 20

Antennae length, especially in males, long but not extending much beyond pronotum

(Fig. 2.4C) ………………………………………………………………………….. 31

20. Male epiproct with a pair of tubercles in the middle (Fig. 2.10B) …...…………….. 21

Male epiproct without a pair of tubercles in the middle (Fig. 2.10A) ..……………. 25

21. Small for the genus (♂TL= 29-32mm; ♀TL= 39-41mm), eyes highly prominent and

protruding. Tegmina narrow and with mottled maculation. Hind femur with two dark

dorsal bands. Taken alive, color green. Museum specimens brown. Endemic to

Central Florida, only found on Florida rosemary Ceratiola ericoides. [Central Florida,

U.S.] ………..………………………………………. S. ceratiola Hubbell & Walker

Body size not small, eyes normal. Tegmina without distinct maculation …...... 22

Figure 2.10: Variation of male epiproct in Schistocerca.

22. Fore and middle femora of males distinctively inflated. Basal eminence of cingulum,

hour-glass shaped, and very broad in the middle …….…………………………….. 23

Fore and middle femora of males not inflated. Basal eminence of cingulum, slightly

or highly constricted in the middle …….……………………………….………..… 24

23. Overall coloration olive green. Upper carina and upper carinula without a row of

small dots. Hind tibiae mostly red, pink, or orange. Posterior margin of abdominal

tergites always without a row of black dots. [Southwestern U.S., including Arizona,

California, Utah, Nevada, and Colorado] ………………..….... S. shoshone (Thomas)

Coloration extremely variable (brown, olive green, or black and yellow). Upper

carina and upper carinula with a row of small dots. Hind tibiae brown, red, or black.

Posterior margin of abdominal tergites always with a row of black dots. [Entire

U.S.] .…………….....…………...……..…………………….…… S. lineata Scudder

24. Overall coloration rusty brown. Pronotum cylindrical, not narrowing toward head.

Head slightly inflated. Dorsal longitudinal stripe usually absent, but can be present.

Male cerci small, quadrate, length about the same as width, distal tip slightly bilobed.

[Eastern U.S., including Connecticut, Florida, Georgia, Massachusetts, New Jersey,

North Carolina, Tennessee, and South Carolina] …………...… S. rubiginosa (Harris)

Overall coloration brownish to slightly olive green. Pronotum narrowing toward head.

Head small and not inflated. Dorsal longitudinal stripe always present. Male cerci

quadrate, length about the same as width, slightly inflated in the middle, distal tip

bilobed. [Eastern U.S.] ………………….…………………..…… S. alutacea (Harris)

25. Overall profile of male subgenital plate slightly longer than wide ………...………. 26

Overall profile of male subgenital plate much longer than wide ………...………… 29

26. Apical lobes of male subgenital plate slightly or much flared outward (Fig. 2.9C) .. 27

Apical lobes of male subgenital plate not flared outward (Fig. 2.9A, B) ...……...… 28

27. Male cerci highly bilobed, and lower apical angle protruding more than upper. Apical

lobes of male subgenital plate only slightly flared outward. Upper carina and upper

carinula of hind femora without a row of small dots. Two distinct dorsal bands on

hind femora with upper half of medial area black. Hind tibiae mostly red, sometime

black or blue. [Southwestern U.S., including Arizona, New Mexico, and Texas, and

Mexico] .…………………………………………………..… S. albolineata (Thomas)

Male cerci not bilobed. Apical lobes of male subgenital plate highly flared outward.

Upper carina and upper carinula of hind femora with a row of small dots. Two dorsal

bands on hind femur present or obliterated to no band. Hind tibiae purple, brown or

black. [Southern U.S. and northern Mexico] ..………………... S. obscura (Fabricius)

28. Male subgenital plate incision shallow. Overall color brownish with olive hue and

bright yellow dorsal stripe running from head to tegmina. Tegmina olive-green

without pattern. [Mexico] ...... … S. cohni Song

Male subgenital plate incision deep. Overall color brown with yellow dorsal stripe

only on head and pronotum. Tegmina sometimes with faded small maculation.

Median carina distinctly raised. [Mexico] ...…..…………….… S. camerata Scudder

29. Male subgenital plate incision U-shaped (Fig. 2.9A). Lophi of epiphallus crested

inward. Lower margin of lateral lobe of pronotum with white marking.

[Trinidad] ………...………………………...……. S. nitens (Thunberg) TRINIDAD

Male subgenital plate incision V-shaped (Fig. 2.9B). Lophi of epiphallus inflated

triangular …..……………………………………………………………………..… 30

30. Overall color olive-green. Tegmina brown without maculation. Dorsal stripe green.

Outer surface of hind femora without maculation. [Costa Rica, Panama] …….………

...... S. crocotaria Scudder

Overall color brown. Tegmina brown with faded maculation. Dorsal stripe yellow.

Outer surface of hind femora with some maculation. [Colombia] …………………...

………………………………………………….. S. nitens (Thunberg) COLOMBIA

31. Posterior margin of metazona of pronotum broadly round (Fig. 2.8A) ……………. 32

Posterior margin of metazona of pronotum angular (Fig. 2.8B, C) …………….….. 33

32. Prosternal process distinctly curved backward, nearly touching thoracic sternum (Fig.

2.6B). Length of tegmina extending only about the length of one pronotum beyond

abdomen (Fig. 2.5B). Lateral stripes on dorsum of pronotum brownish and often

faded. Sculpture on pronotum with small and narrow ridges. [Barbados] ……………..

………………………………………………... S. pallens (Thunberg ) BARBADOS

Prosternal process only slightly curved backward, never touching thoracic sternum

(Fig. 2.6A). Length of tegmina extending distinctly more than the length of one

pronotum beyond abdomen (Fig. 2.5C). Lateral stripes on dorsum of pronotum olive-

green. Lower margin of lateral lobe of pronotum with white marking. [Ecuador,

Peru] …………………………………………… S. pallens (Thunberg ) ECUADOR

33. Posterior margin of metazona of pronotum acutely angular with pointed apex (Fig.

2.8C) ….…………………………………………………………………………….. 34

Posterior margin of metazona of pronotum obtusely angular with round apex (Fig.

2.8B) ….……………………………………………………………………….……. 35

34. Lower margin of lateral lobe of pronotum with white marking. Outer surface of hind

femora with white background numerous black dots. Tegmina with large brown

maculation. Hind tibiae deep wine red with the tip of tibial spines black. [Colombia]

………………………………………………………………….. S. diversipes Hebard

Small for the genus (♂TL= 24-26mm; ♀TL= 36-40mm). Pronotum with two narrow

brown lateral stripes and yellow dorsal stripe running from head to middle of

pronotum. Fore and middle legs light olive-green. Hind tibiae proximally green,

becoming red nearing tarsi. Male cerci narrowing toward apex. [Endemic to

Hispaniola, especially in Dominican Republic] …...… S. quisqueya Rehn & Hebard

35. Male cerci widest near base and narrowing toward apex (Fig. 2.7A) ……………... 36

Male cerci width of apex similar to width of base (Fig. 2.7C) …………………….. 37

36. Upper carina of hind femora numerously serrate. Overall coloration dark green.

Integument of lateral lobe highly rugose. Ovipositor slender and elongated. [Only

known from Brazil] ………………………………………..…… S. braziliensis Dirsh

Upper carina of hind femora regularly serrate. Outer margin of tegmina with white

streak. Lower margin of lateral lobe of pronotum with white marking. Distinct dorsal

stripe on head and pronotum. Hind tibiae usually wine red to purple [West Indies,

Venezuela, Guyana, Surinam, French Guiana, Brazil, Bolivia, Paraguay, Uruguay,

and Argentina] ………...……………………...... … S. flavofasciata (De Geer)

37. Male cerci apex distinctly bilobed (Fig. 2.7D) …………………………………….. 38

Male cerci apex not bilobed (Fig. 2.7A) …………………………………………… 39

38. Overall color variable. Tegmina extending about one pronotum length beyond

abdomen (Fig. 2.5B). Male subgenital plate incision deep and U-shaped. [Colima,

Guerrero, Mexico] …………………………………. S. nitens (Thunberg) COLIMA

Overall color brown. Tegmina extending only slightly beyond abdomen (Fig. 2.5A).

Male subgenital plate incision shallow and V-shaped. [Nuevo Leon, Tamaulipas,

Mexico] …………….……………………….. S. nitens (Thunberg) NUEVO LEON

39. Male cerci apex upper portion modified as a small pointed projection (Fig. 2.7B) .. 40

Male cerci apex without any modification …………...…………………………….. 42

40. Male cerci slightly longer than the width. Apical lobes of male subgenital plate only

slightly flared outward. Overall color variable. Hind wings yellow with smoked color.

[Guatemala] ..………………………………... S. nitens (Thunberg) GUATEMALA

Male cerci distinctly longer than the width ………...…………………………….… 41

41. Ovipositor slender. Overall coloration olive-brown. Endemic to Gorgona Island.

[Gorgona Island, Colombia] ……………………………………..… S. gorgona Dirsh

Tegmina with mottled maculation. Dorsal stripe wide and faintly present. Hind wings

yellow hue. [Vera Cruz, Mexico] ……...……… S. nitens (Thunberg) VERA CRUZ

42. Median carina of pronotum low. Tegmina with dark maculation. Hind margin of

metazona highly constricted. Dorsal stripe distinctly narrowing down in the middle of

pronotum. [St. Kitts, West Indies] ……………………………... S. carribeana (Dirsh)

Median carina of distinctly raised as a ridge ………...…………………………..… 43

43. No distinct markings on lateral prozona. No visible dorsal stripe. [St. Thomas, U.S.

Virgin Islands, West Indies] ………………………………. S. columbina (Thunberg)

Lower margin of lateral lobe of pronotum with white marking. Pronotum with

numerous maculation. [St. Croix, U.S. Virgin Islands, West Indies] …......

…………………………………………………………………...... S. virginica (Dirsh)

2.7 A PARTIAL REVISION OF THE GENUS SCHISTOCERCA

The phylogeny of Schistocerca (see Chapter 3) reveals that many of Dirsh’s (1974) species concepts are paraphyletic and need to be reexamined more carefully.

Synonymized names will need to be resurrected and new species will need to be described. A revision of the entire genus is an enormous undertaking, but a necessary one.

Rather than attempting to revise the genus as a whole, I have been focusing on geographic regions and species occurring on those areas. Here I present taxonomic treatment of seven Schistocerca species and a description of a new species that occur in

Mexico and North America. Six of these species were considered subspecies in Dirsh’s revision. They are S. alutacea, S. rubiginosa, S. lineata, S. shoshone, S. albolineata, and

S. socorro. Detailed morphological examinations along with ecological understanding suggest that these taxa are distinct and valid species. Parts of data presented here have already been published (Song 2004a, 2006).

2.7.1 Materials and Methods

This work was based on the study of specimens (n = 7882) from the following institutions: Academy of Natural Sciences, Philadelphia, PA (ANSP); Brigham Young

University, Arthropod Collection, Provo, UT (BYU); Reginald Chapman’s personal collection (CHAP); Colorado State University Insect Collection, Fort Collins, CO

(CSUC); Illinois Natural History Survey, Champaign, IL (INHS); University of

Wisconsin Insect Research Collection, Madison, WI (IRCW); McGill University, Lyman

Entomological Museum, Ste. Anne de Bellevue, QC, Canada (LEMQ); New Mexico

State University Arthropod Museum, Las Cruces, NM (NMSU); Oklahoma State

University, K.C. Emerson Museum, Stillwater, OK (OSEC); Ohio State University Insect

Collection, Columbus, OH (OSUC); University of Kansas, Snow Entomology Collection,

Lawrence, KS (SEMC); the author’s personal collection (SONG); University of Arkansas

Arthropod Museum, Fayetteville, AR (UARM); University of Michigan Museum of

Zoology, Ann Arbor, MI (UMMZ); University of Minnesota Insect Collection, St. Paul,

MN (UMSP); Universidad Nacional Autónoma de Mexico, Mexico City, Mexico

(UNAM); University of Nebraska State Museum, Lincoln, NE (UNSM); University of

Idaho WFBarr Entomological Museum, Moscow, ID (WFRM). Hubbell’s genitalia collection from UMMZ was also studied. All existing type specimens were examined from the following institutions: Academy of Natural Sciences, Philadelphia, PA (ANSP);

British Museum of Natural History, London, U.K. (BMNH); California Academy of

Sciences, San Francisco, CA (CAS); Harvard University Museum of Comparative

Zoology, Cambridge, MA (MCZ); University of Michigan Museum of Zoology, Ann

Arbor, MI (UMMZ); National Museum of Natural History, Washington, D.C. (USNM).

Male genitalia were extruded by inserting a probe under the epiproct using the technique described by Hubbell (1932). Ovipositors were dissected by making a slit at the distal part of abdomen (Cohn and Cantrall 1974). Genitalia were placed in 10% KOH solution for several hours to dissolve muscles. Cleared genitalia were placed in a vial filled with glycerin, and each genital specimen was given an identification number to associate with the pinned specimens. Illustrations were initially made using a camera lucida mounted on a Wild stereomicroscope and they were then traced in Adobe

Illustrator CS using an optical pen mouse.

An electronic supplement of this study is available as a form of website at www.schistocerca.org/alutacea. This website includes information on each species along with the images of all existing type specimens. It also contains an interactive identification key, a diagnosis, a distribution map, and a list of selected literature. PDF files of the original description are also linked to the website.

2.7.2 Taxonomic treatments

Schistocerca alutacea (Harris, 1841) stat. rev.

(Figs. 2.11A-I)

Acrydium alutaceum Harris, 1841: 139

Schistocerca alutacea (Harris) (Bruner, 1893: 26)

Schistocerca alutacea (Harris) (Henderson, 1942: 101) 39

Schistocerca alutacea (Harris) (Hubbell, 1960: 62)

Schistocerca alutacea alutacea (Harris) (Dirsh, 1974: 194)

Male. Medium size (Total Length = 33 – 40 mm). Integument moderately setose.

Median carina of pronotum distinct but not raised, with shallow sulci. Hind angle of pronotum slightly angular. Lateral lobe of metazona with granules. Pronotum slightly narrowing toward head. Head small and not inflated. Fore and middle femur not inflated.

Epiproct with a pair of tubercles (Fig. 2.11G). Cerci quadrate, length about the same as width, slightly inflated in the middle, distal tip bilobed (Fig. 2.11I). Apical lobes of subgenital plate not outwardly flared, notch U-shaped (Fig. 2.11H). Phallus: Cingulum, surfaces of rami deeply infolded in the middle and highly convex, thus making “basal eminence” appear bilobate and highly constricted in the middle (Fig. 2.11E). Endophallus, basal valves laterally semi-circular, valves of cingulum straight, slightly protruding more than apical valves of aedeagus (Fig. 2.11D). Epiphallus, distance between lophi as long as the length of base of a lophus. Lophi inflated triangular (Fig. 2.11F). Color: Rusty brown to slightly olive green. Head yellow, rusty brown, olive green with faint to strongly dark subocular stripes. A pair of dark brown stripes between eyes from upper half of frontal ridge to occiput, often extending to pronotum. Head, pronotum and tegmina with a yellow dorsal longitudinal stripe. Lateral lobe of pronoza without marking.

Metazona sometimes with small granules. Epimeron without marking. Tegmina brown, sometimes with slightly mottling. Tegminal veins brown. Hind wing with slightly yellow tinge. Posterior margin of abdominal tergites with a row of black dots. Dorsal and ventral

surface of hind femur light brown. Hind femur mostly without dorsal band, but sometimes with a trace. Medial area white sometimes with irregular pattern of black dots.

Carinula with a row of very small black dots. Upper carina without dots. Hind tibia light brown, with tibial spines yellow with black tip.

Female. Much larger than ♂ (TL = 49 – 52 mm). Median carina of pronotum slightly more raised. Otherwise same as male. Ovipositor: Ventral valves long in profile.

Base of dorsal valves distinctively angular. Egg guide slightly narrowing toward apex, slightly curved upward. Pigmentation covering Jannone’s organ distinct, sclerotized linearly in the middle.

Diagnostic Characters. Schistocerca alutacea can be uniquely distinguished based on the highly convex rami of cingulum. It always has a yellow dorsal longitudinal stripe. Front and middle femora of the males are never inflated.

Material Examined. 1223 specimens (♂: 726, ♀: 497) from ANSP, CHAP,

CSUC, INHS, IRCW, LEMQ, OSEC, OSUC, SEMC, UMMZ, and UMSP.

Type Material. NEOTYPE ♂ [MCZ, here designated] Schistocerca alutacea

(Harris, 1841), with labels. “Mass. Martha’s / Vineyard. West / Chop. Aug. 1893,” “West

Chop, Mas. / Aug. 19, 1893,” “PROPERTY / M. C. Z. / Harvard,” “TOPOTYPE /

Schistocerca / alutacea (Harris),” “PLESIALLOTYPE / Schistocerca / alutacea (Harris),”

“PLESIALLOTYPE / See Hubbell / 1960: 62,” “NEOTYPE / Schistocerca / alutacea

(Harris) / H. Song, 2003.” (Fig. 2.11A, B)

Distribution. Examined specimens were collected from Alabama, Arkansas,

Connecticut, Florida, Georgia, Maryland, Michigan, Mississippi, New Jersey, New York,

North Carolina, Oklahoma, Pennsylvania, South Carolina, Tennessee, Virginia, and

Wisconsin. This species is distributed mainly in the eastern U.S.

Biology. Schistocerca alutacea prefers shrubby, moist to wet situations, including bogs, swamps, marshes, and thickets bordering mesic forests (Hubbell 1960). Besides wet habitats, Squitier and Capinera (2002a) found that the species also prefers dry, sand hill. In Florida, nymphs occur starting in May, and adults can be frequently found in July and August (Squitier and Capinera 2002a, b).

Taxonomic Discussion. Although Hubbell (1960) considered the genital characters to be important for the Alutacea Group, Dirsh (1974) apparently overlooked these characters. His revision contained only two drawings of the cingulum of S. alutacea alutacea, and none for the other subspecies (Dirsh, 1974: 195). These drawings, however, failed to show the distinct “basal eminence” that Hubbell (1960) emphasized, making it difficult to validate Dirsh’s taxonomic concept. I examined many phallic complexes of S. alutacea from various collecting localities, as well as the entire genitalia collection used by Hubbell, and I conclude that Hubbell’s (1960) characterization was correct, thus reviving the taxonomic status to a valid species.

The type specimen in the collection of the Boston Society of Natural History was destroyed (Hubbell 1960: 62). In the original description, Harris (1841: 139) did not specify the sex of the type specimen, but the original measurement (length: 1 ¾ inch; wingspan: 3 inches) suggests that it was a female. Hubbell designated a plesiallotype, a male specimen topotypic of the type specimen, which, however, is not recognized by the

International Code of Zoological Nomenclature (ICZN). Vickery and Kevan (1983)

considered this plesiallotype as a neotype, but no formal action was taken. Therefore, I hereby designate Hubbell’s plesiallotype as a neotype.

Schistocerca rubiginosa (Harris, 1862) stat. rev.

(Figs. 2.11A-I)

Acridium rubiginosum Harris, in Scudder, 1862: 467

Schistocerca rubiginosa (Harris) (Hubbell, 1960: 66)

Schistocerca alutacea rubiginosa (Scudder) (Dirsh, 1974: 198)

Male. Medium size (TL = 34 – 39 mm). Integument without long setae. Median carina of pronotum distinct but not raised, with shallow sulci. Hind angle of pronotum slightly angular. Pronotum without pattern or granule. Anterior end of pronotum broad, and head slightly inflated. Overall pronotum cylindrical. Fore and middle femur not inflated. Epiproct with a pair of tubercles (Fig. 2.12G). Cerci small, quadrate, length about the same as width, distal tip slightly bilobed (Fig. 2.12I). Apical lobes of subgenital plate not outwardly flared, notch U-shaped (Fig. 2.12H). Phallus: Cingulum, surfaces of rami infolded in the middle and moderately convex, thus making “basal eminence” appear slightly bilobate and somewhat constricted in the middle (Fig. 2.12E).

Endophallus, basal valves laterally semi-circular, valves of cingulum straight, slightly protruding more than apical valves of aedeagus (Fig. 2.12D). Epiphallus, distance between lophi as long as the length of base of a lophus (Fig. 2.12F). Lophi inflated triangular. Color: Rusty brown. Head rusty brown with numerous small black dots, with

faintly brown subocular stripes. A pair of rows of black dots between eyes from fastigium to occiput. Dorsal longitudinal stripe rarely present. Epimeron without pattern. Antennae rusty brown. Tegmina rusty brown with slight mottling. Tegminal veins rusty brown.

Hind wing with yellow tinge. Posterior margin of abdominal tergites with a row of black dots. Dorsal and ventral surface of hind femur rusty brown. No dorsal band on hind femur. Medial area white to light brown with irregular pattern of black dots. Upper and lower carinula with a row of black dots. Upper carina with a row of dots. Hind tibia light to rusty brown, with tibial spines yellow with black tip.

Female. Much larger than ♂ (TL = 49 – 54 mm). Median carina of pronotum slightly more raised. Otherwise same as male. Ovipositor: Ventral valves long in profile.

Base of dorsal valves distinctively angular. Egg guide slightly narrowing toward apex, slightly curved upward. Pigmentation covering Jannone’s organ distinct, faintly sclerotized linearly in the middle.

Diagnostic Characters. Schistocerca rubiginosa can be distinguished by the slightly inflated head and the cylindrical pronotum. It also has the smallest male cerci compared to other species in the group. Overall coloration is almost always rusty brown, and most southern populations do not have a dorsal longitudinal stripe. This species is different from S. alutacea in the morphology of the rami of cingulum, which is only slightly convex.

Material Examined. 1292 specimens (♂: 795, ♀: 497) from ANSP, CSUC,

IRCW, LEMQ, OSEC, SEMC, UMMZ, UMSP, UNSM, and WFRM.

Type Material. NEOTYPE ♂ [UMMZ] Schistocerca rubiginosa (Harris, 1862),

with labels. “BEAUFORT CO. S.C. / 1.1 mi N. Limehouse / (US17) Aug. 20, ’47 / 2

T.H.Hubbell,” “PLESIALLOTYPE / Schistocerca / rubiginosa (Harris),”

“PLESIALLOTYPE / See Hubbell / 1960:66,” “Head and sub- / genital plate fig. /

Hubbell 1960 pl.14,” “NEOTYPE / Schistocerca / rubiginosa (Harris) / H. Song, 2003.”

(Fig. 2.12A, B)

Distribution. Examined specimens were collected mainly from the eastern U.S.

including Connecticut, Florida, Georgia, Massachusetts, New Jersey, North Carolina,

Tennessee, and South Carolina.

Biology. Schistocerca rubiginosa prefers xeric to xeromesic habitats, especially

sandy soil (Hubbell 1960). In Florida, it is sympatric with S. alutacea and has a similar

seasonal phenology (Song, unpublished data).

Taxonomic Discussion. Schistocerca rubiginosa was considered a synonym of S.

alutacea for almost sixty years until Hubbell (1960) revised the group because of Rehn’s

(1901, 1902b) erroneous observation. On the basis of morphological characters, I

conclude that Hubbell’s (1960) characterization was correct, thus reviving the taxonomic

status to a valid species.

As in S. alutacea, Hubbell (1960) designated a plesiallotype, which is not recognized by the Code. Vickery and Kevan (1983) considered this as a neotype, but no formal action was taken. The rusty brown color of the species has sometimes caused misidentification, especially with female specimens, and even Dirsh erroneously identified it as either S. damnifica or S. lineata. Squitier and Capinera (2002a, b) did not

recognize the species and considered it as a synonym of S. alutacea. In order to prevent further taxonomic confusion, I think it is necessary to designate a neotype. Thus, I hereby designate Hubbell’s plesiallotype as a neotype.

Schistocerca lineata Scudder, 1899 stat. rev.

(Figs. 2.13A-I)

Acridium emarginatum Uhler, manuscript

Acridium emarginatum, nomen nudum Dodge, 1872: 15

Acridium emarginatum Scudder, 1872: 250

Schistocerca lineata Scudder, 1899: 465

Schistocerca scudderi Bruner, 1906: 676, unnecessary replacement name

Schistocerca lineata Scudder (Hubbell, 1960: 71)

Schistocerca emarginata (Scudder) (Vickery and Kevan, 1964: 1555; 1983: 725)

Schistocerca alutacea lineata (Scudder) (Dirsh, 1974: 204)

Male. Medium size (TL = 32 – 50 mm). Integument sparsely setose. Median carina of pronotum distinct and slightly raised, with distinct sulci. Hind angle of pronotum broadly angular. Lateral lobe of metazona mostly with granules. Pronotum slightly narrowing to head, and head medium size. Fore and middle femur highly inflated.

Epiproct with a pair of tubercles (Fig. 2.13G). Cerci quadrate, length about the same as width, slightly inflated in the middle, distal tip bilobed (Fig. 2.13I). Lower apical angle of cerci protruding slightly more than upper. Apical lobes of subgenital plate not outwardly

flared, notch U-shaped (Fig. 2.13H). Phallus: Cingulum, surfaces or rami not infolded and sinuate, thus making “basal eminence” appear hour-glass shaped, and broad in the middle (Fig. 2.13E). Endophallus, basal valves laterally semi-circular, valves of cingulum straight, slightly protruding more than apical valves of aedeagus (Fig. 2.13D). Epiphallus, distance between lophi as long as the length of base of a lophus (Fig. 2.13F). Lophi inflated triangular. Color: Hubbell (1960) analyzed the color variation of the species, and here I categorize it into four distinguishable ecotypes: typical, brown, aposematic, and olive green. Here I characterize the typical form, and the differences among ecotypes are shown in Table 2.3. Head light to greenish brown with brown subocular stripes. A pair of light brown stripes between eyes from upper half of frontal ridge to occiput. Head, pronotum, and tegmina usually with a light yellow dorsal longitudinal stripe. Lateral lobe of prozona without distinct marking. Metazona with small light yellow granules.

Epimeron with small yellow granules. Tegmina light brown, with tegminal veins light brown. Hind wing with slightly yellow tinge. Posterior margin of abdominal tergites with a row of black dots. Dorsal and ventral surface of hind femur light brown. Hind femur without dorsal bands. Medial area light brown. Upper and lower carinula with a row of black dots. Upper carina without dots. Hind tibia black especially on ventral side, with tibial spines yellow with black tip.

Female. Much larger than ♂ (TL = 45 – 69 mm). Median carina of pronotum slightly more raised. Fore and middle femur not inflated. Otherwise same as male.

Ovipositor: Ventral valves long in profile. Base of dorsal valves distinctively angular.

Egg guide highly narrowing toward apex, highly curved upward. Pigmentation covering

Jannone’s organ distinct, sclerotized linearly in the middle.

Diagnostic Characters. Schistocerca lineata is the most polymorphic species in

the Alutacea Group, and each local population is highly variable in coloration (Table 2.3),

and perhaps host preference. Therefore, it is very difficult to characterize the species.

However, the species can be distinguished by the phallic morphology, in which the “basal

eminence” resembles an hour-glass and is broad in the middle. This genital character was

first introduced by Hubbell (1960). The species also possesses the inflated front and

middle femora in males, which are also found in S. shoshone. However, S. lineata can

always be distinguished from S. shoshone by the presence of the black dots on the

abdominal tergites.

Material Examined. 3151 specimens (♂: 1656, ♀: 1495) from ANSP, BYU,

CSUC, INHS, IRCW, LEMQ, NMSU, OSEC, OSUC, SEMC, UARM, UMMZ, UMSP,

UNSM, and WFRM.

Type Material: LECTOTYPE ♂ [ANSP] Schistocerca lineata Scudder, 1899, with labels. “Barbour Co. / Kas. Cragin,” “Schist. / lineata / Scudder’s / Type, 1899,” “Ex

Coll.L.Bruner / S. lineata Sc. / Proc. Am. Ac. A.&S. / Vol. XXXIV p. 466. / Hebard

Collection,” “Schistocerca / lineata Sc. / TYPE H86,” “Sch. alutacea / lineata Sc. / Lect. /

V.M.Dirsh det., 1972.” (Fig. 2.13A-B).

Distribution. This is the most widely distributed species of Schistocerca in North

America. Examined specimens were collected from Arizona, Arkansas, Connecticut,

Colorado, Delaware, Georgia, Idaho, Illinois, Indiana, Iowa, Kansas, Kentucky,

Massachusetts, Maryland, Michigan, Minnesota, Mississippi, Nebraska, New Jersey,

New Hampshire, New Mexico, New York, North Dakota, Ohio, Oklahoma, Rhode Island,

South Dakota, Texas, Utah, Virginia, and Wisconsin. Its distribution extends north to

Manitoba and Alberta, Canada.

Biology. Schistocerca lineata is abundant in sandy areas but is also frequently found in other habitats. Detailed account of its habitat association is found in Hubbell

(1960: 43-48). Criddle (1932) reported that the development takes 39 days from the time of hatching. He also noted that this species has five nymphal instars. Sword and colleagues have shown that the Texas population has a density-dependent polyphenism in nymphal instar, mediated by the host preference (Sword 1999, Sword and Dopman 1999,

Sword 2002, Dopman et al. 2002).

Taxonomic Discussion. Here, I designate four ecotypes of S. lineata. Dopman et al. (2002) found that the populations associated with either Ptelea or Rubus form a monophyletic clade using 16S rRNA and 12S rRNA regions of the mitochondrial DNA.

Although their study demonstrated the lack of gene flow between two host-associated populations, I am hesitant to create a new taxonomic concept at this moment, because not much is known about the populations in the midwestern and eastern U.S. Highly conspicuous individuals from the southwestern states are more robust in form and slightly larger than non-conspicuous ones. In terms of coloration, specimens from Great Lake regions are brown and resemble S. rubiginosa. Aposematic specimens are never found in northern states, although the polymorphism of dorsal longitudinal stripe exists. In western

Ecotypes Dorsal Distribution Head Epimeron Tegmina Hind femur Hind tibia stripe

Typical Great Plains light to greenish light yellow with small yellow light brown with light brown without dorsal black especially on ventral region brown with granules light brown veins bands; carinula with a row of side, with tibial spines yellow brown subocular black dots; upper carina with black tip stripes without dots

Brown Great Lake brown with faint absent without pattern light brown with brown without dorsal bands; light brown, with tibial spines regions subocular stripes slight mottling, carinula with a row of black yellow with black tip with brown veins dots; upper carina sometimes with a row of dots

Aposematic Southwestern yellow with black yellow entirely yellow yellowish brown, light brown to yellow with black especially on ventral U.S. subocular stripes without distinct sometimes with distinct black dorsal bands; side, with tibial spines yellow

50 marking, but slight mottling, carinula with/without a row with black tip sulci between with brown veins of very small black dots;

episternum and upper carina mostly without epimeron black. dots

Olive green Western U.S. olive green with bright with small yellow light brown to olive olive green sometimes with pinkish red, with tibial spines dark green yellow granules green, with lemon two dorsal bands; carinula yellow with black tip subocular stripes yellow veins sometimes with a row of black dots, but usually absent; upper carina sometimes with dots, but usually absent.

Table 2.3: Four ecotypes of S. lineata and their color variation

states, specimens become almost indistinguishable from S. shoshone except on the basis of markings on abdominal tergites. These specimens have olive green hue, which is rarely observed in either southwestern or eastern populations. In all cases, however, crucial morphological characters, such as phallic complex and inflated male femurs, are identical.

Schistocerca lineata has sometimes been called S. emarginata by a few authors

(Vickery and Kevan 1964, 1967, Cantrall 1968, Vickery and Kevan 1983, Sword 1999,

Sword and Dopman 1999, Dopman et al. 2002, Sword 2002) because of an invalid action by Vickery and Kevan (1964) to strictly adhere to the Principle of Priority. The taxonomic history of S. lineata is rather complex and is partially summarized in Hubbell

(1960) and Vickery and Kevan (1964, 1983). Here, I explain how confusion arose and argue why S. lineata is the valid name.

Uhler characterized Acridium emarginatum only in a manuscript and never published it, and Dodge (1871) was the first one to use this name as nomen nudum when he listed several grasshopper species collected in Nebraska. Scudder (1872: 250) characterized A. emarginatum and attributed this name to Uhler but noted that the name came only from the manuscript. Through this paper, Scudder inadvertently established himself as author of A. emarginatum. The type specimen of A. emarginatum was lost, and a neotype was never designated (Hubbell 1960). In 1899, Scudder incorrectly synonymized A. emarginatum with S. alutacea and described a new species, S. lineata.

From 1899 to 1960, S. lineata was used continuously and unambiguously with two exceptions (Kellogg 1905, Osborn 1939). Kellogg (1905) used S. emarginata under a

figure copied from Lugger (1898), and the figure alone is difficult to associate with the name because many Schistocerca species are externally similar. Osborn (1939) merely listed several names of injurious grasshoppers, and the name alone is difficult to associate with a taxonomic concept. In 1960, Hubbell published a thorough taxonomic revision on

S. lineata, S. alutacea and S. rubiginosa. He acknowledged that lineata is a strict synonym of emarginata but nevertheless retained S. lineata because of the prevailing use.

His revision effectively reversed the precedence under Articles 23.9 of the ICZN, and S. lineata became a valid name as nomen protectum. In 1964, Vickery and Kevan reinstated

S. emarginata based on the Principle of Priority. Their justification for this action came from the prevailing usage of the name between 1872 and 1898. However, A. emarginatum was the only available name during that period, and S. lineata was described in 1899. Because the name emarginata was already suppressed by Hubbell

(1960) in accordance with the Code, Vickery and Kevan (1964) could not have simply reinstated the name. In order to do so, they should have submitted a proposal to the

Commission, which they never did. Between 1964 and 1974, S. emarginata was used once by Vickery and Kevan (1967). In the complete revision of the genus, Dirsh (1974) agreed with Hubbell's (1960) use of lineata, even though his discussion of synonymy indicated that emarginata has priority. Although he cited Vickery and Kevan (1964),

Dirsh (1974) did not mention their action in the discussion perhaps because he rejected their proposal. Vickery and Kevan (1983), apparently troubled by Dirsh’s revision, dismissed the revision in its entirety. Once again, they reinstated S. emarginata. Since

then, S. emarginata (=lineata) was used in a series of ecological studies by Sword and

Dopman (1999), Sword (1999), Dopman et al. (2002), and Sword (2002).

It is true that the name Acridium emarginatum (or S. emarginata) is older, and

thus taking precedence to S. lineata. Scudder’s (1872) description is sufficient to make him an author but insufficient to distinguish A. emarginatum from other Schistocerca

species. With inadequate characterization and no type specimen available, the name itself

has a dubious status. Hubbell (1960) discovered this older name but nevertheless used S.

lineata in his revision because of the prevailing use of the latter name. Because the name

emarginatum had not been used in the primary zoological literature for more than 50

years, and because the later name lineata was being used prevalently, Hubbell (1960) did

not consider it necessary to keep the antiquated name emarginatum. In order to reverse

the precedence based on the 50-year rule, one needs to apply to the Commission only if it

is proposed from 1961. Interestingly, Hubbell's paper was published on 29 December

1960. Therefore, one can assume that, under the rules in existence at that time, Hubbell

did not need the action of the Commission to declare a name as nomen oblitum. Hubbell

(1960) appreciated the objectives of the Code, which are to promote stability and not

upset a long-accepted name in its accustomed meaning. Vickery and Kevan’s (1964)

subsequent actions were thus erroneous and irrelevant. In order to maintain the stability

and universality of a taxonomic name, partially hindered by Vickery and Kevan’s (1964)

action, I argue that Hubbell (1960)’s initial action was correct, effectively making S.

lineata a valid name as nomen protectum.

Schistocerca shoshone (Thomas, 1873) stat. rev.

(Figs. 2.14A-I)

Acridium shoshone Thomas, 1873: 295

Schistocerca venusta Scudder, 1899: 467 (Hebard, 1935: 299)

Schistocerca obliquata Scudder, 1899: 470 (Hebard, 1932: 279)

Schistocerca shoshone (Thomas) (Henderson, 1942: 99)

Schistocerca alutacea shoshone (Thomas) (Dirsh, 1974: 200)

Male. Medium size (TL = 39 – 44 mm). Integument without long setae. Median

carina of pronotum distinct and slightly raised, with distinct sulci. Metazona with

numerous small granules. Hind angle of pronotum slightly angular. Fore and middle

femur highly inflated. Epiproct with a pair of tubercles (Fig. 2.14G). Cerci quadrate,

length about the same as width, slightly inflated in the middle, distal tip bilobed (Fig.

2.14I). Apical lobes of subgenital plate not outwardly flared, notch U-shaped ((Fig.

2.14H). Phallus: Cingulum, surfaces or rami not infolded and sinuate, thus making “basal eminence” appear hour-glass shaped, and very broad in the middle (Fig. 2.14E).

Endophallus, basal valves laterally semi-circular, valves of cingulum straight, slightly protruding more than apical valves of aedeagus (Fig. 2.14D). Epiphallus, distance between lophi as long as the length of base of a lophus (Fig. 2.14F). Lophi inflated triangular. Color: Mostly light olive green, occasionally lemon yellow or slight brown.

Two forms can be distinguished based on the presence of dorsal stripe. Unstriped form:

Head olive green to yellow with dark olive green to brown subocular stripe. No pair of

stripes between eyes, and no dorsal longitudinal stripe on head and pronotum. Lateral lobe of pronoza without marking. Metazona sometimes with small granules. Antennae yellow. No pattern on epimeron. Tegmina light olive green to yellow without any pattern.

Tegminal veins light lemon-yellow to brown. Hind wing with slightly yellow tinge.

Posterior margin of abdominal tergites without a row of black dots. Dorsal and ventral hind femur yellow to olive green. Medial area white without any pattern. Hind femur without dorsal bands. Carinula without pattern. Hind tibia pink, orange to red, with tibial spines yellow with black tip. Striped form: Head light olive green to yellow with faint dark olive green to brown subocular stripes. A pair of olive green stripes between eyes from upper half of frontal ridge to occiput. Head, pronotum and tegmina with a yellow dorsal longitudinal stripe. Lateral lobe of prozona with no apparent marking, and lower half of prozona and entire metazona with yellow granules. Second epimeron sometimes with yellow marking. Antennae, tegmina, hind wing, abdomen, and hind femur same as unstriped form.

Female. Much larger than ♂ (TL = 53 – 63 mm). Median carina of pronotum distinctly more raised. Fore and middle femur not inflated. Otherwise same as male.

Ovipositor: Ventral valves short in profile. Base of dorsal valves distinctively angular.

Egg guide highly narrowing toward apex, highly curved upward. Pigmentation covering

Jannone’s organ distinct, sclerotized linearly at the base.

Diagnostic Characters. Schistocerca shoshone can be easily distinguished by the absence of black markings on the abdominal tergites. These black markings appear to be the universal color pattern in Schistocerca species, but S. shoshone always lacks this

pattern. This species is closely related to S. lineata, because both share the similar phallic morphology and inflated fore and middle femora in males.

Material Examined. 991 specimens (♂: 498, ♀: 493) from ANSP, BYU, CSUC,

INHS, IRCW, LEMQ, OSEC, OSUC, NMSU, SEMC, UARM, UMMZ, UMSP, UNSM, and WFRM.

Type Material. 1. NEOTYPE ♂ lost. “Utah, Cache County, Logan” (Dirsh, 1965:

40) I was unable to find the neotype at ANSP. 2. LECTOTYPE ♂ [ANSP, dissected phallus stored in a vial] Schistocerca venusta Scudder, 1899, with labels. “Indio, Calif. /

July 9, 1897,” “Schist. / venusta / Scudder’s / Type, 1899,” “S.H.Scudder / Coll.,” “Type

/ 1829,” “Lecto / type / R.&H. 1912,” “venusta / Scudder.” 3. LECTOTYPE ♂ [ANSP, dissected phallus stored in a vial] Schistocerca obliquata Scudder, 1899, with labels.

“San Jose / del cabo / Mexico,” “Schist. / obliquata / Scudder’s / Type, 1899,” “Ex

Coll.L.Bruner / S. obliquata Sc. / Proc. Am. Ac. A.&S. / Vol. XXXIV p. 471 / Hebard

Collection,” “Schistocerca / obliquata Sc. / TYPE H393,” “Lectotype.” (Fig. 2.14A, B)

Distribution. Schistocerca shoshone occurs in the western U.S. including

Arizona, California, Colorado, Idaho, Nevada, New Mexico, Oregon, Texas, and Utah.

Biology. It is present in both riparian and desert habitats and prefers various woody plants (Sword and Chapman 1994). In Utah, it is frequently found in cornfields or other tall growing vegetation (Henderson 1942). This species is univoltine, and adults occur in June and July (Chapman et al. 1995).

Taxonomic Discussion. Schistocerca shoshone has both striped and unstriped forms that appear to be associated with certain host plants (Chapman et al. 1995).

Because these two forms can apparently interbreed (Chapman and Sword, unpublished data), I hereby call these ecotypes of a single species.

Dirsh’s (1974) redescription of S. alutacea shoshone is suspicious, because it states that the anterior femora of males are slightly or not at all inflated. All type specimens as well as all the examined specimens possess this character, and usually the inflation is even more obvious than in S. lineata. Therefore, I argue that Dirsh’s synonymy is erroneous, and I hereby revive the taxonomic status to a valid species.

Schistocerca albolineata (Thomas, 1875) stat. rev., syn. nov.

(Figs. 2.15A-I)

Acridium albolineata Thomas, 1875: 897

Schistocerca mexicana Scudder, 1899: 468 (Hebard, 1932: 281)

Schistocerca insignis Hebard, 1932: 279 (syn. nov.)

Schistocerca chinatiensis Tinkham, 1948: 607 (Dirsh, 1974: 209)

Schistocerca alutacea insignis (Hebard) (Dirsh, 1974: 212)

Schistocerca alutacea albolineata (Thomas) (Dirsh, 1974: 209)

Male. Medium size (TL = 39 – 43 mm). Integument setose, with long setae on pronotum, thorax, cerci, and subgenital plate. Median carina slightly raised, with distinct sulci. Metazona with numerous small granules. Hind angle of pronotum slightly angular.

Width of metazona slightly wider than width of prozona. Fore and middle femur not inflated. A pair of tubercles on epiproct absent (Fig. 2.15G). Cerci quadrate, length 1.5

times longer than width, and bilobed (Fig. 2.15I). Lower apical angle of cerci protruding

more than upper. Apical lobes of subgenital plate slightly flared outward, notch round V-

shaped (Fig. 2.15H). Phallus: Cingulum, surfaces of rami deeply infolded in the middle

and highly convex, thus making “basal eminence” appear slightly bilobate and somewhat

constricted in the middle (Fig. 2.15E). Endophallus, basal valves ventral angle protruding

more than dorsal, valves of cingulum club-shaped and curved downward, protruding

more than apical valves of aedeagus (Fig. 2.15D). Epiphallus, distance between lophi

longer than the length of base of a lophus (Fig. 2.15F). Lophi lamelliform. Color: Olive brown. Head yellowish to olive brown with dark-brown to black subocular stripes. A pair of olive brown stripes between eyes from upper half of frontal ridge to occiput. Head and pronotum with a bright yellow dorsal longitudinal stripe, sometimes extending to tegmina.

Epimeron often distinctly yellow. Antennae yellow. Lateral lobe of pronotum sometimes with a yellow rectangular spot. Tegmina uniformly olive brown, with veins light-brown to yellow. Hind wing with yellow tinge. Posterior margin of abdominal tergites with a row of black dots. Dorsal and ventral surface of hind femur yellow, with two distinct black dorsal bands. Posterior tip of hind femur black. Outer face of hind femur, upper half of medial area sometimes black and lower half white. Carinula and carina without spots. Hind tibia mostly red or orange-red, but occasionally black or blue, with tibial spines yellow with black tip.

Female. Much larger than ♂ (TL = 49 – 61 mm). Median carina of pronotum more raised. Otherwise same as male. Ovipositor: Ventral valves short in profile. Base of

dorsal valves not angular. Egg guide not narrowing toward apex, slightly curved upward.

Pigmentation covering Jannone’s organ distinct, sclerotized circularly in the middle.

Diagnostic Characters. Males of S. albolineata posses elongated cerci with the

lower apical angle protruding more than upper, which is unique among the species in the

Alutacea Group. The following characters can also be found in S. obscura, but two

species are never sympatric. They also have an elongated subgenital plate that is slightly

flared outward. The epiproct lacks a pair of tubercles. The epiphallus is wide in the

bridge between lophi, and the apical valve of cingulum is clubbed. Most known

specimens have a bright yellow dorsal longitudinal stripe, two black dorsal bands on hind

femur, and red hind tibia.

Material Examined. 240 specimens (♂: 140, ♀: 100) from ANSP, BYU, NMSU,

and UMMZ.

Type Material: 1. NEOTYPE ♂ [ANSP] Acridium albolineata Thomas, 1875,

with labels. “Ajo, Pima Co. Ariz. / ab.1800 ft. (R&H) / IX, 18, 1922,” “Schistocerca /

alutacea albolineata Thom. / V.M. Dirsh det. 1971,” “NEO-TYPE.” (Fig. 2.15A, B) 2.

HOLOTYPE ♂ [ANSP, dissected phallus stored in a vial] Schistocerca mexicana

Scudder, 1899, with labels. “Sinaola, Mex. / KOELS. J. BEHRENS,” “Schist, / mexicana

/ Scudder’s / Type, 1899,” “S.H.Scudder / Coll.,” “Type 1830,” “Holotype,”

“HOLOTYPE / Sch. mexicana / Scudder, 1899. / V.M.Dirsh 1964.” 3. HOLOTYPE ♂

[ANSP, dissected phallus stored in a vial] Schistocerca insignis Hebard, 1932, with labels.

“Guadalajara Jalisco / 9.19 1903. Mex. / J.F. McClendon,” “Schistocerca / insignis /

Hebard, / TYPE 5509.” 4. ALLOTYPE ♀ [ANSP] Schistocerca insignis Hebard, 1932,

with labels. “Guadalajara Jalisco / 9.18 1903. Mex. / J.F. McClendon,” “Schistocerca /

insignis / Hebard, / Allotype ♀.” 5. HOLOTYPE ♀ [CAS] Schistocerca chinatiensis

Tinkham, 1948, with labels. “Chinatis / 3 mi W of / Schafter / 19 Oct 46,” “Type /

Schistocerca / chinatiensis / msp. 1947 Tink.,” “ERNEST ROBERT TINKHAM /

COLLECTION – 1988 / DONATED TO THE CALIFORNIA / ACADEMY OF

SCIENCES / by Marion Blair Tinkham,” “California Academy / of Sciences / Type no.

16968.”

Distribution. Schistocerca albolineata occurs in the southwestern U.S. including

Arizona, New Mexico, and Texas. Its distribution extends south to Jalisco, Mexico.

Biology. Schistocerca albolineata prefers dry mountainous habitats. It usually

feeds on various woody plants, and adults can be found in early September (Song,

unpublished data). On Tucson Mountain, Arizona, I found a population of emerging

adults feeding on foothill Palo Verde Cercidium microphyllum (Torrey) Rose & Johnston

(Fabaceae). Howard (1995) found that the Arizona populations preferred cotton

Gossypium hirsutum L. (Malvaceae).

Taxonomic Discussion. Although S. albolineata is the most homogeneous

species in the Alutacea Group in terms of coloration, examination of a large number of

specimens revealed that the color variation indeed exists especially in hind tibia. Color

variation in Schistocerca is well known (Hubbell 1960), and there seems to be the

southward darkening progression in the species. Schistocerca chinatiensis Tinkham,

described based on a local population from the Chinati Mountains, Texas, has slightly

darker coloration and black tibia. Schistocerca insignis Hebard, only known from eight

specimens from Guadalajara, Mexico, is even darker in general coloration and has black

tibia. Hubbell (1960) suggested that it is possible for S. albolineata, S. chinatiensis, and S.

insignis to be the same polytypic species, and I hypothesize that this is indeed a case here.

All known specimens of the populations from Arizona, New Mexico, Texas, and Mexico possess the invariable diagnostic characters. Therefore, I accept Dirsh’s (1974) synonymy of S. chinatiensis, and I hereby synonymize S. insignis under S. albolineata on the basis of morphological similarity. I also argue that Dirsh’s synonymy overlooked the distinct morphological characters in S. albolineata, and I hereby revive the taxonomic status to a

valid species.

Schistocerca obscura (Fabricius, 1798)

(Figs. 2.16A-I)

Gryllus obscurus Fabricius, 1798: 194

Schistocerca obscura (Fabricius) (Scudder, 1899: 465)

Schistocerca obscura (Fabricius) (Dirsh, 1974: 181)

Male. Medium size (TL = 38 – 46 mm). Integument highly setose. Median carina of pronotum distinct and slightly raised, with distinct sulci. Hind angle of pronotum slightly angular. Width of metazona slightly wider than width of prozona. Fore and middle femur not inflated. A pair of tubercles on epiproct absent (Fig. 2.16G). Cerci quadrate, length twice longer than width, not bilobed (Fig. 2.16I). Apical lobes of subgenital plate strongly flared outward, notch deep (Fig. 2.16H). Phallus: Cingulum,

surfaces of rami deeply infolded in the middle and highly convex, thus making “basal

eminence” appear bilobate and highly constricted in the middle (Fig. 2.16E). Endophallus,

basal valves ventral angle protruding more than dorsal, valves of cingulum club-shaped

and curved downward, protruding more than apical valves of aedeagus (Fig. 2.16D).

Epiphallus, distance between lophi much longer than the length of base of a lophus (Fig.

2.16F). Lophi lamelliform. Color. Olive green to liver color. Head yellowish to liver color with dark-brown to black subocular stripes. A pair of dark brown stripes between eyes from upper half of frontal ridge to occiput. Head and pronotum with a bright yellow dorsal longitudinal stripe, often extending to tegmina. Upper half of epimeron often distinctly yellow. Antennae yellow to orange. Lateral lobe of pronotum without pattern.

Lower half of lateral prozona and entire metazona with yellow granules. Tegmina uniformly purplish liver color except for the extending dorsal longitudinal stripe.

Tegminal veins same color as tegmina. Hind wing transparent with yellow tinge.

Posterior margin of abdominal tergites with a row of black dots. Dorsal and ventral surface of hind femur yellowish olive to liver color, with two black dorsal bands which sometimes reduced to a trace. Medial area mostly white. Upper and lower carinula on outer face with a row of small black dots (Fig. x). Hind tibia brown to black with tibial spines yellow with black tip.

Female. Much larger than ♂ (TL = 52 – 70 mm). Median carina of pronotum

more raised. Otherwise same as male. Ovipositor: Ventral valves long in profile. Base of

dorsal valves distinctively angular. Egg guide slightly narrowing toward apex, ventral

portion highly sinuate. Pigmentation covering Jannone’s organ distinct, sclerotized circularly in the middle and patterns present on top.

Diagnostic Characters. Schistocerca obscura can be easily distinguished from other species in the Alutacea Group on the basis of the highly flared male subgenital plate

(Fig. 2.16H). Male cerci are elongated with length twice longer than width, and not bilobed (Fig. 2.16I). It also has the epiproct lacking a pair of tubercles (Fig. 2.16G), the wide epiphallus, and the clubbed apical valve of cingulum, which it shares with S. albolineata.

Material Examined. 985 specimens (♂: 495, ♀: 490) from ANSP, BYU, CHAP,

CSUC, INHS, IRCW, LEMQ, NMSU, OSEC, OSUC, SEMC, UARM, UMMZ, UMSP,

UNSM, and WFRM.

Type Material. 1. NEOTYPE ♂ [BMNH] Schistocerca obscura (Fabricius,

1798), with labels, “Allen Co Ks / Ele. 062,15 / R.H. Beamer,” “NEO- / TYPE,”

“Brit.Mus. / 1925-208,” “Schistocerca / obscura / (Fabr),” “Schistocerca / obscura (Fabr,

1798) / V.M. Dirsh det., 1974.” (Fig. 2.16A, B).

Distribution. Examined specimens were collected from Alabama, Arkansas,

Florida, Georgia, Kansas, Louisiana, Maryland, Mississippi, Missouri, Nebraska, New

Mexico, North Carolina, Oklahoma, South Carolina, Tennessee, Texas, and Virginia. Its distribution extends south to Tamaulipas, Mexico.

Biology. Duck (1944) documented the bionomics of S. obscura after the grasshopper outbreak in Oklahoma. It prefers to feed on woody plants such as American elm and cotton. The species has five nymphal instars and emerges as an adult

approximately 50 d after hatching. Populations in central and south Texas are commonly

associated with hackberry, Celtis laevigata Willd. (Ulmaceae) (Chapman and Sword, unpublished data). Its preferred habitat is fields and open woodlands (Capinera et al.

2001). Emerging adults were collected in August in Florida (Song, unpublished data).

Taxonomic Discussion. Dirsh (1974) treated S. obscura as a species distinct from

S. alutacea based on the shape of subgenital plate (Fig. 2.16H). However, the phylogenetic analysis (in Chapter 3) suggests that S. obscura is sister to S. albolineata indicating the paraphyly of S. alutacea sensu Dirsh (1974). No taxonomic action is taken for S. obscura in this revision, but this species is now considered to be included in the

Alutacea Group.

Schistocerca cohni n. sp.

(Figs. 2.17A-I)

Male. Medium size (total length = 39.2 ± 3.15 mm; hind femur length = 20.13 ±

1.87 mm; pronotum length = 6.97 ± 0.6 mm (n = 6)). Antennae much longer than the combined length of head and pronotum. Integument highly setose. Setae on pronotum, sternum, abdomen and femora long. Median carina of pronotum distinct and slightly raised, but not constricted. Sulci distinct, but not deep. Sculpting pattern of dorsal surface of prozona papillulate. Lateral lobes of prozona slightly wrinkled. Hind angle of pronotum slightly obtuse-angular. Small granules present on pronotum and thorax.

Anterior and posterior portion of pronotum similar width. Tegmina slightly extending

beyond the tip of abdomen. Cerci quadrate, with length twice longer than width (Fig.

2.17I). Apical tip of cercus slightly bilobed, with lower part extruding more than upper.

Furcula small and rectangular. Epiproct with a pair of tubercle absent (Fig. 2.17G).

Subgenital plate with round apex (Fig. 2.17H). Apical lobes of subgenital plate not

outwardly flared, with U-shaped notch. Color: Rusty brown to slightly deep olive green.

Head rusty brown to deep olive green with faint to strongly dark subocular stripes. A pair of dark brown stripes between eyes from upper half of frontal ridge to occiput, often extending to pronotum. Head, pronotum and tegmina with a bright yellow dorsal longitudinal stripe. Lateral lobes of pronoza without marking. Metazona with granules.

Epimeron without marking. Tegmina uniformly brown. Tegminal veins brown. Hind wings with slightly yellow tinge. Posterior margin of abdominal tergites with a row of black dots. Dorsal and ventral surface of hind femora light brown to olive green. Hind femora mostly without dorsal bands, but sometimes with a trace. Medial area white.

Carinula with a row of very small black dots. Upper and lower carina without dots. Hind tibiae dark reddish brown to black, with tibial spines yellow with a black tip. Phallus:

Cingulum, surfaces of rami slightly infolded in the middle and somewhat convex, thus

making “basal eminence” appear slightly bilobate and somewhat constricted in the

middle (Fig. 2.17E). Endophallus, basal valves ventral angle protruding more than dorsal,

valves of cingulum club-shaped and curved downward, protruding more than apical

valves of aedeagus (Fig. 2.17D). Epiphallus, distance between lophi longer than the

length of base of a lophus (Fig. 2.17F). Lophi lamelliform.

Female. Much larger than male (total length = 59.37 ± 3.27 mm; hind femur

length = 29 ± 1.91 mm; pronotum length = 10.83 ± 0.6 mm (n = 3)). Median carina of

pronotum more raised. Otherwise same as male.

Type Material. Holotype male. Mexico: Guerrero: 9 rd. mi. NE. Taxco (1.7 rd.

mi. SW. Acuitlapan) 5700 ft. 17 Sep 1959, I.J.Cantrall & T.J.Cohn, #137 (deposited at

University of Michigan Museum of Zoology) (Fig. 2.17A, B).

Collecting Localities. Eight paratypes ( ♂: 5, ♀: 3) from UMMZ (Fig. 3A).

Mexico: Guerrero: 9 rd. mi. NE. Taxco (1.7 rd. mi. SW. Acuitlapan) 5700 ft. 17 Sep

1959, I.J.Cantrall & T.J.Cohn, #137 (2 ♂ + 1 ♀); Jalisco: 12.4 mi. N. Barra de Navidad

(on Hwy 80) 5 Oct 1970, T.J.Cohn & J.W.Cohn, #43 (2 ♂); Oaxaca: 2.5 mi. E. La

Ventosa (12 mi. NE Juchitán) 150 ft. 13 Sep 1959, I.J.Cantrall & T.J.Cohn, #107 (1 ♂);

Puebla: 3 mi. SE. Petlalcingo 4900ft. 15 Sep 1959, I.J.Cantrall & T.J.Cohn, #126 (1 ♂);

San Luis Potosi: El Pujal. 100 ft. 18 Jul 1939, R.Haag (1 ♀).

Diagnostic Characters. Schistocerca cohni has a bright yellow dorsal stripe from head to tegmina. Antennae are much longer than the combined length of head and pronotum, especially in males. It can be easily distinguished from the species in the

Alutacea Group on the basis of male epiproct, which lacks a pair of tubercles and the male subgenital plate which is not flared outwardly.

Biology. Not much is known about S. cohni. The specimens were collected from open woodlands with a thick growth of medium-height weeds. Nothing is known about its feeding preferences, but the overall ecology and behavior are likely to be similar to other species in the Alutacea Group.

Taxonomic Discussion. This species was initially discovered during a

morphological phylogenetic analysis of Schistocerca. The specimens were identified by

Dirsh as S. obscura (Fabricius), but they clearly lacked a characteristic subgenital plate of

S. obscura. Further examinations on both internal and external morphological structures revealed that it was a new species overlooked by previous taxonomists. The collecting localities indicate that this species is widespread in various habitats. This species is phylogenetically closely related to the species in the Alutacea Group (see Chapter 3).

Schistocerca cohni is superficially similar to S. lineata, but has a distinctly different epiproct where a pair of tubercles is absent. This loss character is also found in S. albolineata and S. obscura, but S. cohni is distinct from these two species because the apical lobes of subgenital plate are not flared outwardly.

Etymology. Named in honor of the eminent orthopterist, Dr. Theodore J. Cohn, who collected the type series.

Schistocerca socorro (Dirsh, 1974) n. stat. and just. emend.

(Figs. 2.18A-I)

Schistocerca americana socoro Dirsh, 1974: 59

Male. Medium size (total length = 49.95 ± 3.03 mm; hind femur length = 21.98 ±

1.80 mm; pronotum length = 8.83 ± 0.41 mm (n = 10)). Antennae slightly longer than the combined length of head and pronotum. Integument highly setose. Setae on pronotum, sternum, abdomen and femora short. Median carina of pronotum distinct and slightly

raised, and slightly constricted. Sulci distinct, but not deep. Dorsal surface of prozona sculpting pattern very finely granulous with faint ridges. Lateral lobes of prozona faintly papillulate and wrinkled. Hind angle of pronotum slightly obtuse-angular. Granules absent. Pronotum width narrowing anteriorly. Tegmina extending beyond the tip of abdomen. Cerci slightly narrowing toward apex, with length about the same as width (Fig.

2.18I). Apical tip of cercus almost round, sometimes faintly bilobed, with upper and lower part forming elongated “3” shape. Furcula broad and sinuate, narrowing at the base.

Epiproct with a pair of tubercle absent (Fig. 2.18G). Subgenital plate with smoothly angular and broad apex (Fig. 2.18H). Subgenital plate notch V-shaped. Color: Overall dark brown. Dorsal stripe absent. Head brown with two subocular stripes, white anterior to dark brown. Two dark brown stripes converging on dorsal portion of head from behind eyes to occiput. Lateral lobes of pronoza with a faintly white horizontal streak. Marginal ridge of pronotum cream color. Epimeron brown without distinct marking. Tegmina semitransparent with distinct dark patches. Tegminal veins dark brown. Hind wings smoky. Posterior margin of abdominal tergites with a row of black dots. Dorsal and ventral surface of hind femora light brown. Hind femora without dorsal bands. Bottom half of medial area white. Dark stripe along upper carinula. Upper and lower carina without dots. Hind tibiae light brown, with tibial spines brown with black tip. Phallus:

Cingulum, surfaces of rami not infolded in the middle and sinuate, thus making “basal eminence” appear hourglass-shaped and broad in the middle (Fig. 2.18E). Endophallus, basal valves ventral angle protruding slightly more than dorsal, overall semicircular (Fig.

2.18D). Valves of cingulum pointed, protruding more than apical valves of aedeagus.

Epiphallus, distance between lophi shorter than the length of base of a lophus (Fig.

2.18F). Lophi right-triangular.

Female. Much larger than male (total length = 67.5 ± 4.58 mm; hind femur length

= 30 ± 2.60 mm; pronotum length = 15.83 ± 3.75 mm (n = 3)). Similar to male.

Type Material. 1. PARATYPE ♀ [BMNH] with labels, “Socorro I. /

RevillaGigedo Is. / III-27-32,” “M Willows Jr. / Collector,” “Templeton / Crocker /

Exped. 1932,” “Para- / type,” “From Collection / Calif Acad Sci,” “USNM,” “Brit. Mus. /

1981-203,” “Ph,” “32,” “Schistocerca am. / socoro Dirsh ♀ / V.M.Dirsh det., 1974.” 2.

PARATYPE ♂ [BMNH] with labels, “Socorro Id. / Brthwte Bay / May 7, 1925,”

“H.H.Keifer / Collector,” “Para- / type,” “From Collection / Calif Acad Sci,” “USNM,”

“Brit. Mus. / 1981-203,” “Ph,” “Schistocerca am. / socoro Dirsh ♂ / V.M.Dirsh det.,

1974.” 3. PARATYPE ♀ [USNM] with labels, “Socorro Id. / Brthwte Bay / May 7,

1925,” “H.H.Keifer / Collector,” “Para- / type,” “USNM,” “31,” “SCHISTOCERCA /

AMERICANA / SOCORO DIRSH / V.M. Dirsh det., 1971.” 4. PARATYPE ♂ [USNM]

with labels, “Socorro Id. / Brthwte Bay / May 7, 1925,” “H.H.Keifer / Collector,” “Para- /

type,” “From Collection / Calif Acad Sci,” “USNM,” “22,” “SCHISTOCERCA /

AMERICANA / SOCORO DIRSH / V.M. Dirsh det., 1971.” (Fig. 2.18A, B)

Additional Material. 13 specimens (10 ♂, 3 ♀) collected during CONABIO-

sponsored trip to Socorro Island (October 17-November 5, 2004). Mexico: Colima: Islas

Revillagigedo, Isla Socorro. Road to al Volcan in the forest N18o45.494’ W110 o 57.557’

618 m, 20-X-2004; 40 min. hike up from Senna field (El Paradero) N18o46.331’ W110 o

57.402’ 1716 ft, 25-X-2004; Playa Norte N18o51.527’ W110 o 59.249’ 153 ft, 27-X-2004;

Road to Playa Norte N18o46.283’ W110 o 55.595’ 345 m, 26-X-2004; Road to Playa

Norte N18o47.172’ W110 o 56.031’ 361 m, 26-X-2004; Road to Playa Norte N18o48.011’

W110 o 56.269’ 360 m, 29-X-2004.

Diagnostic Characters. Schistocerca socorro is overall dark brown with no

dorsal stripe. This species is only found on Socorro Island and seems to prefer a forest

habitat. Two migrant Schistocerca species, S. piceifrons and S. nitens, co-occur on the

island, and an angular hind margin of pronotum of the endemic species is a useful

character to distinguish it from S. piceifrons, which has a round hind margin of pronotum.

Dark patches on tegmina can be used to distinguish the endemic species from S. nitens

which has a mottled pattern on tegmina.

Distribution. This species is endemic to Socorro Island, Mexico. Socorro Island

is the largest of four islands comprising the Islas Revillagigedo, located about 480

kilometers southwest of Baja California, Mexico. The type series of S. socorro was

collected from Braithwaite Bay, a mixed habitat located at lower elevations of the island.

However, a migrant locust species, S. piceifrons, currently inhabits at Braithwaite Bay in

large numbers. Recently collected material of S. socorro is all from higher elevations, in

the forest in Mt. Evermann and in Playa Norte. This could indicate that the endemic

species is limited to higher elevations perhaps due to competitive exclusion.

Biology. Schistocerca socorro is an arboreal species that prefers to feed on native

herbaceous plants. It is behaviorally sedentary and is a strong flyer. In early 2003, a

group of Mexican officials and researchers working on Socorro Island reported a

significant locust outbreak. The locust species was later identified as S. piceifrons, which

apparently colonized the island from the mainland Mexico at unknown time. Because of the voracity of the locusts, especially to the native flora, CONABIO (Comisión Nacional para el Conocimiento y Uso de la Biodiversidad) issued a year-long ecological study of the impact of S. piceifrons to the island biota. CONABIO was particularly concerned with the negative impact of the locusts to the endemic Schistocerca species, then known as S. americana socoro. This endemic species had not been collected since 1925. In

October 2004, a group of researchers from Universidad Nacional Autónoma de Mexico

(UNAM) and I visited Socorro Island and found several sustainable populations of the endemic species. We also identified several ecological differences between S. socorro and S. piceifrons. Although there are areas where two species co-occur, we found that the endemic species is mostly confined to the forested areas in higher elevations, whereas the locust species occur in disturbed areas in lower elevations. We do not know how the endemic species interacts with the locust species, and a continuous monitoring of the population dynamics is desperately needed.

Taxonomic Discussion. Dirsh (1974) originally described Schistocerca americana socoro based on a series collected in 1925 by Hartford H. Keifer of the

California Academy of Sciences. He described it as a subspecies of S. americana, because he reasoned that patterns on hind femora and tegmina were indicative of an affinity to the nominal species, but these characters have since shown to be highly variable and taxonomically unreliable. In a phylogenetic analysis (Chapter 3), I show that that all the species in the Americana Complex sensu Harvey (1981) are grouped by a round hind margin of pronotum, which S. socorro lacks. Thus, here I argue for a

taxonomic status change from a subspecies to a valid species. Dirsh (1974) also used a subspecific epithet socoro despite the fact that the type label clearly states the correct spelling of the island. Schistocerca socorro is the only endemic grasshopper species on

Socorro Island, and I feel it is important for it to reflect the correct name of the island.

Thus, I propose a justified emendation of the species epithet from socoro to socorro.

Figure 2.11: Schistocerca alutacea. A. lateral view of type; B. dorsal view of type; C. cingulum; D. endophallus; E. basal eminence of zygoma; F. epiphallus; G. male epiproct; H. male subgenital plate; I. male cercus; (Ac: arch of cingulum; Anc: ancora; Ap: apical valve of aedeagus; Apd: apodemes of cingulum; Bp: basal valve of aedeagus; Br: bridge of epiphallus; Cv: valve of cingulum; Gpr: gonopore process; Lp: lophus; Rm: rami of cingulum; Zyg: zygoma.

Figure 2.12: Schistocerca rubiginosa. A. lateral view of type; B. dorsal view of type; C. cingulum; D. endophallus; E. basal eminence of zygoma; F. epiphallus; G. male epiproct; H. male subgenital plate; I. male cercus. 74

Figure 2.13: Schistocerca lineata. A. lateral view of type; B. dorsal view of type; C. cingulum; D. endophallus; E. basal eminence of zygoma; F. epiphallus; G. male epiproct; H. male subgenital plate; I. male cercus. 75

Figure 2.14: Schistocerca shoshone. A. lateral view of type; B. dorsal view of type; C. cingulum; D. endophallus; E. basal eminence of zygoma; F. epiphallus; G. male epiproct; H. male subgenital plate; I. male cercus. 76

Figure 2.15: Schistocerca albolineata. A. lateral view of type; B. dorsal view of type; C. cingulum; D. endophallus; E. basal eminence of zygoma; F. epiphallus; G. male epiproct; H. male subgenital plate; I. male cercus. 77

Figure 2.16: Schistocerca obscura. A. lateral view of type; B. dorsal view of type; C. cingulum; D. endophallus; E. basal eminence of zygoma; F. epiphallus; G. male epiproct; H. male subgenital plate; I. male cercus. 78

Figure 2.17: Schistocerca cohni. A. lateral view of type; B. dorsal view of type; C. cingulum; D. endophallus; E. basal eminence of zygoma; F. epiphallus; G. male epiproct; H. male subgenital plate; I. male cercus. 79

Figure 2.18: Schistocerca socorro. A. lateral view of type; B. dorsal view of type; C. cingulum; D. endophallus; E. basal eminence of zygoma; F. epiphallus; G. male epiproct; H. male subgenital plate; I. male cercus. 80

2.8 TAXON DESCRIPTION

It is beyond the scope of this work to revise the genus Schistocerca in its entirety.

In Chapter 3, I present a phylogenetic hypothesis of Schistocerca based on morphological

characters. During the course of present study, I discovered several phylogenetically

useful characters, which may be useful in the future revision of the genus. Below I

present taxon descriptions for 46 Schistocerca species that were included in the phylogenetic analysis. These are not meant to be a formal taxonomic description, but rather a summary of phylogenetically important characters. A taxon description for

Halmenus robustus was included because it is considered to be a sister genus to

Schistocerca.

Halmenus robustus [Endemic to Galápagos Islands, Ecuador; Isla Genovesa, Islas

Rábida, Santa Cruz, Santiago]

Antennae: length visibly longer than the combined length of head and pronotum, especially in males; Pronotum: prozona not constricted in the middle when viewed

dorsally; posterior margin of metazona broadly round; median carina low; sculpture

pattern on dorsum of prozona ridged; lateral ridges on dorsum of prozona created by

irregular shaped punctures and overall faded; sculpture pattern of lateral lobe of prozona

smooth; granules on dorsum of pronotum absent; sculpture pattern of metazona

numerously pitted with depressions created by ridges small and ridges narrow;

Prosternal process: cylindrical and distinctly conical; Tegmina: brachypterous; Male

anterior femora: inflated; Hind femora: upper carina serration absent; upper carinula

granules miniscule; Male cerci: simple triangular with pointed apex; Male epiproct: ridges forming lateral lobes not reaching to the middle; a pair of elongate ridges along the basal-lateral sides present; Male subgenital plate: shape of apex divided into two lobes; lobe weakly present and not protruding; Female subgenital plate: lateral lobes not projecting forward; Phallic complex: overall size small; Epiphallus: projection of lateral lobes absent or only a trace of projection; length of bridge between lophi normal; lophi shape widely triangular with pointed apex; lophi angle relative to bridge projecting nearly parallel to bridge; lophi inflated triangular; Ectophallic sclerite: lateral overall profile mid-projection protruding as a small lobe; Cingulum: overall shape of lateral apodeme

U-shaped when viewed dorsally; overall shape of zygoma and rami distinct with membranous zygoma; short and rami completely closed with less membrane at apex; shape of basal eminence round; Endophallus: valve of penis elongated and very narrow and thin in its entirety absent; apical valve two valves not fused; apical valve of cingulum narrowing toward apex and very short.

Schistocerca americana [Southern U.S. from Texas to Florida]

shaped punctures; sculpture faded near median carina; sculpture pattern of lateral lobe of prozona smooth; granules on dorsum of pronotum absent; sculpture pattern of metazona numerously pitted with depressions created by ridges small and ridges narrow;

Prosternal process: cylindrical and straight or slightly curved backward; Tegmina: length extending more than the length of one pronotum beyond abdomen; pattern large brown maculation; area above costa waxy white pattern present; background colorless;

Male anterior femora: inflated; Hind femora: upper carina regularly serrated; upper carinula granules miniscule; Male cerci: quadrate; widest near base and narrowing toward apex; shape of apex only lightly bilobed; Male epiproct: ridges forming lateral lobes not reaching to the middle; a pair of elongate ridges along the basal-lateral sides absent; a pair of tubercles in the middle absent; Male subgenital plate: shape of apex divided into two lobes; distinctly protruding as lobes; overall profile slightly longer than wide; flare on lobes absent; depth of incision shallow; shape of incision V-shaped;

Female subgenital plate: lateral lobes not projecting forward; Phallic complex: overall size small; Epiphallus: projection of lateral lobes absent or only a trace of projection; length of bridge between lophi normal; lophi shape widely triangular with pointed apex; lophi angle relative to bridge projecting nearly parallel to bridge; lophi shape simply right triangular; Ectophallic sclerite: lateral overall profile mid-projection protruding as a small lobe; Cingulum: overall shape of lateral apodeme U-shaped when viewed dorsally; overall shape of zygoma and rami distinct with membranous zygoma; short and rami completely closed with less membrane at apex; shape of basal eminence round;

Endophallus: valve of penis elongated and very narrow and thin in its entirety absent;

apical valve two valves not fused; apical valve of cingulum narrowing toward apex and distinctly protruding from cingulum.

Schistocerca piceifrons [Mexico south of the Tropic of Cancer and Central America, with a subspecies S. piceifrons peruviana in Peru and southern Ecuador]

Antennae: length as long as the combined length of head and pronotum, especially in males; Pronotum: prozona visibly constricted in the middle when viewed dorsally; posterior margin of metazona broadly round; dorsal color pattern clearly defined lateral stripe with broad dorsal stripe; median carina low; sculpture pattern on dorsum of prozona ridged; lateral ridges on dorsum of prozona created by irregular shaped punctures; sculpture faded near median carina; sculpture pattern of lateral lobe of prozona smooth; granules on dorsum of pronotum absent; sculpture pattern of metazona numerously pitted with depressions created by ridges small and ridges narrow;

wide; flare on lobes absent; depth of incision shallow; shape of incision V-shaped;

Female subgenital plate: lateral lobes not projecting forward; Phallic complex: overall

size small; Epiphallus: projection of lateral lobes absent or only a trace of projection;

length of bridge between lophi normal; lophi shape widely triangular with pointed apex;

lophi angle relative to bridge projecting nearly parallel to bridge; lophi shape simply right

triangular; Ectophallic sclerite: lateral overall profile mid-projection protruding as a

small lobe; Cingulum: overall shape of lateral apodeme U-shaped when viewed dorsally;

overall shape of zygoma and rami distinct with membranous zygoma; short and rami

completely closed with less membrane at apex; shape of basal eminence round;

Endophallus: valve of penis elongated and very narrow and thin in its entirety absent; apical valve two valves not fused; apical valve of cingulum narrowing toward apex and slightly protruding from cingulum.

Schistocerca cancellata [Chile, Argentina, Bolivia, Paraguay, Uruguay and southern

Brazil]

Antennae: length as long as the combined length of head and pronotum, especially in males; Pronotum: prozona visibly constricted in the middle when viewed dorsally;

posterior margin of metazona broadly round; dorsal color pattern clearly defined lateral

stripe with broad dorsal stripe; median carina raised as a ridge; sculpture pattern on

dorsum of prozona ridged; lateral ridges on dorsum of prozona created by irregular

shaped punctures; sculpture faded near median carina; sculpture pattern of lateral lobe of

prozona smooth; granules on dorsum of pronotum numerously present; sculpture pattern

of metazona numerously pitted with depressions created by ridges small and ridges

narrow; Prosternal process: cylindrical and straight or slightly curved backward;

Tegmina: length extending more than the length of one pronotum beyond abdomen;

pattern large brown maculation; area above costa waxy white pattern present; background

colorless; Male anterior femora: inflated; Hind femora: upper carina regularly serrated; upper carinula granules miniscule; Male cerci: quadrate; widest near base and narrowing toward apex; shape of apex distinctly bilobed; Male epiproct: ridges forming lateral

lobes not reaching to the middle; a pair of elongate ridges along the basal-lateral sides

absent; a pair of tubercles in the middle absent; Male subgenital plate: shape of apex

divided into two lobes; distinctly protruding as lobes; overall profile slightly longer than

wide; flare on lobes absent; depth of incision shallow; shape of incision V-shaped;

Female subgenital plate: lateral lobes not projecting forward; Phallic complex: overall

size small; Epiphallus: projection of lateral lobes absent or only a trace of projection;

length of bridge between lophi normal; lophi shape widely triangular with pointed apex;

lophi angle relative to bridge projecting nearly parallel to bridge; lophi shape right

triangular with inner side inflated; Ectophallic sclerite: lateral overall profile mid-

projection protruding as a small lobe; Cingulum: overall shape of lateral apodeme U-

shaped when viewed dorsally; overall shape of zygoma and rami distinct with

membranous zygoma; short and rami completely closed with less membrane at apex;

shape of basal eminence rounded rectangular; Endophallus: valve of penis elongated and

very narrow and thin in its entirety absent; apical valve two valves not fused; apical valve

of cingulum narrowing toward apex and distinctly protruding from cingulum.

Schistocerca gregaria [Northern Africa and Middle East, with a subspecies, S. gregaria

flaviventris found in the Republic of South Africa]

Antennae: length as long as the combined length of head and pronotum, especially in males; Pronotum: prozona visibly constricted in the middle when viewed dorsally;

posterior margin of metazona broadly round; dorsal color pattern clearly defined lateral

stripe with broad dorsal stripe; median carina raised as a ridge; sculpture pattern on

dorsum of prozona ridged; lateral ridges on dorsum of prozona created by irregular

shaped punctures; sculpture faded near median carina; sculpture pattern of lateral lobe of

prozona smooth; granules on dorsum of pronotum absent; sculpture pattern of metazona

numerously pitted with depressions created by ridges small and ridges narrow;

Prosternal process: cylindrical and straight or slightly curved backward; Tegmina:

length extending more than the length of one pronotum beyond abdomen; pattern large

brown maculation; area above costa waxy white pattern present; background colorless;

Male anterior femora: inflated; Hind femora: upper carina regularly serrated; upper

carinula granules miniscule; Male cerci: quadrate; widest near base and narrowing

toward apex; shape of apex only lightly bilobed; Male epiproct: ridges forming lateral