Er. .. . up? 7.0 an n 4%.. 53...”. .3. £315... v a: .3". D.) .. . . II. .. 1 try Lucia. {.1- «wr..{ H? :1. u . V .5; 1 (HS. . 55$. (1.: é...) flaw “rs“..édu 3%..va um. .. ...qum.ioqmu..z ; n. I. «T... ahefik... 2.2.1. . 3...”... u $138. 2 $1.93? I «Janna. «grahfiuhzfl 3 ..... .i .21: H; s . 5 , .. a“??? .. . gun. a . ‘ to.» 1 .ba. ix .. Iv «2.33 Ff... £3 2.3.. . . x .3. .5 3.3.4. K any}. ‘ u.....\ «1.-.». 1.1.1“)? THESIS 3 ITY LIBFIIARIE ||l||ll|l|llllllllllllllllllllllllllll Illllll 3 1293 01555 9465 This is to certify that the dissertation entitled Systematic Studies of Rhagoletis and Related Genera (Diptera: Tephritidae) presented by John Jenkins has been accepted towards fulfillment of the requirements for Ph . D . degree in Zoology [KM/m Major professor 036% 2,. ma MSU is an Affirmative Action/Equal Opportunity Institution 0- 12771 LIBRARY Michigan State University PLACE II RETURN BOX to moon this chockout from your mood. TO AVOID FINES totum on or befoto duo duo. DATE DUE DATE DUE DATE DUE MSU Is An Aflirmotivo Action/Equal Oppotttmity Instituion W i 1 SYSTEMATIC STUDIES OF RHAGOLETIS AND RELATED GENERA (DIPTERA: TEPHRITIDAE) By John Jenkins A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Zoology 1996 ABSTRACT SYSTEMATIC STUDIES OF HHAGOLETISAND RELATED GENERA (DIPTERA: TEPHRITIDAE) By John Jenkins Two traditional sources of taxonomic characters. male genitalia and wing patterns, were examined in detail, and relationships among Rhagoletis and 16 related genera were analyzed. The genitalia of 278 males in 90 species was examined. A detailed description of the male genitalia based on these examinations is given. A ground plan for the phallus is proposed, and homology of genital structures is discussed. Elements of banded wing patterns are identified using structural landmarks instead of their relative position on the wing. A model of wing pattern evolution is presented, and a transformation series for wing patterns in Rhagoletis is given. A phylogenetic analysis of 50 species of Rhagoletis and 38 species in 17 other trypetine genera was performed. During the character analysis, 247 characters were examined. resulting in 91,942 recorded observations. Characters used in the cladistic analysis are detailed, and the use of polymorphisms as cladistic characters is discussed. Results of the cladistic analysis indicate that Hhagoletis is not monophyletic; that the subtribe Carpomyina is monophyletic and the subtribe Trypetina is paraphyletic; and that previously unplaced trypetines may be closely related to the Trypetina. To my Folks I would like to acknowledge the assistance of my graduate committee, G. L. Bush (Major Professor), D. J. Hall, K. L. Klompatens, J. J. Smith, and E. T. VanTassell. l gratefully acknowledge financial assistance from the Department of Zoology, Michigan State University, for Teaching Assistantships and support for travel to Washington, D. C.; from G. L. Bush for Research Assistantships and research materials; the College of Natural Science, Michigan State University, for 3 Continuing Scholar Fellowship; and the Entomological Society of America for an R. E. Snodgrass Memorial Fund grant. The following individuals and institutions-loaned specimens used during this study. P. Arnaud Jr., California Academy of Sciences; D. Azuma, Philadelphia Academy of Natural Science; A. O. Bachmann, Museo Argentine de Ciencias Naturales 'Bemardino Rivadavia" e Instituto Nacional de lnvestigacion de Ias Ciencias Naturales (Buenos Aires, Argentina); J. K. Barnes, New York State Museum; 8. B. Barrios, Museo Nacional de Historia Natural del Paraguay (San Lorenzo, Paraguay); 8. H. Berlocher, University of Illinois; R. L. Blinn, North Carolina State University; H. D. Blocker, Kansas State University; T. Bierbaum, Michigan State University; M. D. Bowers, University of Colorado; M. Brancucci, Naturhistorisches Museum Basel (Switzerland); R. W. Brooks and G. Byers, University of Kansas; S. G. Cannings, University of British Columbia; C. Carlton, University of Arkansas; J. Chainey, The Natural History Museum (British Museum) (London, England); D. Chandler, University of New Hampshire; J. B. Chapin, Louisiana State University; J. A. Chemsak, University of California, Berkeley; P. G. Clausen, University of Minnesota; R. Contreras-Lichtenberg, Naturhistorisches Museum iv Wien (Austria); J. M. Cumming, Agriculture Canada (Ontario); M. Daccordi and L. Sorbini, Museo Civico di Storia Naturale, Verona (Italy): R. Danielsson, Lund University (Sweden); J. C. Deeming and A. H. Kirk-Spriggs, National Museum of Wales (Cardiff, Wales); K. Dirlbek, Praha (Czech Republic); J. L. Feder, University of Notre Dame; W. A. Foster, University of Cambridge (England); A. Freidberg, Tel Aviv University (Israel); P. H. Freytag, University of Kentucky; 8. I. Frommer, University of California, Riverside; D. Furth, C. T. Graham, and C. Vogt, Harvard University; P. Grootaert, Institut Royal des Sciences Naturelles de Belgique, Brussels (Belgium); H-Y. Han, Pennsylvania State University; W. J. Hanson, Utah State University; J. Harrington and S. Krauth, University of Wisconsin; V. Hernandez Ortiz, Institute De Ecologia, A. C. (Veracruz, Mexico); M. W. Heyn, Clemson University; C. L. Hogue, Los Angeles County Museum of Natural History; J. ltomies, University of Oulu (Finland); M. Ivie, and C. Seibert, Montana State University; L. G. Jensen, University of Bergen (Norway); J. Jezek, Narodni Muzeum v Ptaze, Praha (Czech Republic); C. Johnson, University of Manchester (England); U. Kallweit, Staatliches Museum fur Tierkunde Dresden (GermanY): E. Kierych, Polish Academy of Science, Warszawa (Poland); L. S. Kimsey, University of California, Davis; 8. Kondratieff, Colorado State University; V. A. Komeyev, Ukrainian Academy of Sciences (Ukraine); M. Kosztarab and M. Rhoades, Virginia Polytechnic Institute; J. D. Latin, and J. A. DiGiulio, Oregon State University; J. K. Liebherr, Cornell University; K. C. Kim, J. Luhman, and S. R. Moulton, ll, Pennsylvania State University; L. Lyneborg, University of Copenhagen (Denmark); S. Marshall, University of Guelph (Ontario); 8. Mascherini, Sezione del Museo di Storia Naturale, Firenze (Italy); K. C. McGiffen, Illinois Natural History Survey; 8. A. McPheron, Pennsylvania State University; J. E. McPherson, Southern Illinois University; F. Menzel, Projektgruppe Entomologie in der Fachhochschule Eberswalde (Germany); F. W. Merickel and J. B Johnson, University of Idaho; B. Merz, ETH- Zentrum, Zfirich (Switzerland); 8. E. Miller and K. Arakaki, Bishop Museum; A. F. Newton, Jr. and P. P.. Parillo, Field Museum of Natural History; A. L. Norrbom, U. S. National Museum, Washington, 0.0.; H. D. Nunez, Escuela Nacional de Ciencias Forestales (Honduras); M. F. O'Brien, University of Michigan; I. Okéli, Slovenské Narodné Muzeum, Bratislava (Czech Republic); C. Olson, University of Arizona; E. A. Osgood, University of Maine; T. M. Peters, University of Massachusetts; J-G. Pilon, University of Montreal (Canada); A. Provonsha and D. Bloodgood, Purdue University; 8. C. Radcliffe, University of Nebraska; M. Reilly, Glasgow University (Scotland); V. Richter, Zoological Institute of the Academy of Sciences, Leningrad (Russia); E. G. Riley, Texas A and M University; R. Ripa, Instituto de Investigaciones Agropecuarias, La Cruz (Chile); R. A. Ronderos, Universidad Nacional de La Plata (Argentina); C. Schaefer, University of Connecticut; T. L. Schiefer, Mississippi State University; J. T. Schulz and E. U. Balslaugh, Jr., North Dakota State University; H. Schumann, Universitat zu Berlin (Germany): S. “R. Shaw, University of Wyoming; D. Shpeley, University of Alberta; R. Sites, University of Missouri; C. L. Smith, University of Georgia; G. Stahls, University of Helsinki (Finland); J. Stark, American Museum of Natural History; F. Stehr, Michigan State University; N. A. Straw, Forestry Commission, Wrecclesham Farnham Surrey (England); D. W. Tallamy, University of Delaware; D. Torres and J. Kochalka, Cuidad Universaria, San Lorenzo (Paraguay); C. A. Triplehorn, Ohio State University; H. Ulrich, Zoologisches Forschungsinstitut und Museum “Alexander Koenig" (Bonn, Germany); K. Valley, Pennsylvania Department of Agriculture; V. Vallo, Slovak Academy of Sciences (Slovakia); T. von Proschwitz, Naturhistoriska Museet, Goteborg (Sweden); X-j. Wang, Academia Sinica (China); H. Weems and G. Steck, Florida State Collection of Arthropods; R. L. Westcott, Oregon Department of Agriculture; I. M. White, CAB. International (London, England); A. E. Whittington, (8. James, and K. J. Duxbury, Natal Museum (South Africa); G. B. Wiggins, Royal Ontario Museum (Canada); C. W. Young, Carnegie Museum of Natural History (Pittsburg); R. 8. Zack, Washington State University; W. Zimmermann, Museen der Stadt Gotha (Germany); R. A. Zucchi, vi Universidade de 850 Paulo (Brasil); H. Zwolfer and M. Leclaire, Universitaet Bayreuth (Germany). ‘ I thank the following people for their help and encouragement: M. Case, M. A. Condon, D. Dale, M. Donoghue, J. Foland, W. Kelly, V. A. Korneyev and E. P. Kameneva, A. L. Norrbom, D. Olmstead, A. Peters, D. Steane, D. O. Straney, W. J. Turner, J. Wilterding, Yue Ming and J. Bedoyan, and J. Zablotny. And then there was my Jude... vii TABLEOFCONTENTS LIST OF TABLES ................................................................................................................. i x UST OF FIGURES ................................................................................................................ x INTRODUCTION ................................................................................................................... 1 CHAPTER 1 MALE GENITALIA IN THE TRYPETINI (DIPTERA: TEPHRITIDAE) ................. -- - -5 Materials and Methods ................................................................................................ 6 Description ....................................................... -. ......................................................... 7 Discussion ................................................................................................................. 1 4 CHAPTER 2 A HEURISTIC MODEL OF WING PATTERN EVOLUTION IN THE TRYPEI'INI (DIPTERA: TEPHRITIDAE) ............................................................................................... 2 3 Evolution of Banded Wing Patterns in the Trypetini ................................................ 2 4 Mechanisms of Wing Pattern Formation .................................................................. 3 2 CHAPTER 3 PI-IYLOGENEI’IC ANALYSIS OF RHAGOLEWS AND RELATED GENERA (DIPTERA: TEPHRITIDAE) .................................................................................................................. 3 9 Materials and Methods .............................................................................................. 4 0 Results and Discussion .............................................................................................. 4 2 SUMMARY ......................................................................................................................... 8 6 APPENDIX ......................................................................................................................... 8 9 UST OF REFERENCES ..................................................................................................... 1 7 5 viii UST OF TABLES Table 1. Specimens examined .......................................................................................... 8 9 Table 2. Comparison of terminology used in naming wing bands ................................... 9 2 Table 3. Classification of genera included in this study (after Foote et al., 1993) ...... 9 3 Table 4. Distribution and larval hosts of specimens examined ...................................... 9 4 Table 5. Character-state matrix used in cladistic analysis ........................................... 9 7 Table 6. Characters occurring in single species .......................................................... 1 1 1 Table 7. Characters used in cladistic analysis .............................................................. 1 1 2 Table 8. Characters not included in the cladistic analysis ........................................... 1 1 7 Table 9. Leg coloration by segment, excluding tarsi .................................................... 12 0 Table 10. Species with tergal patterns matching the medial pattern systema ............ 123 Table 11. Species with tergal patterns matching the sublateral pattern systema ...... 126 UST OF FIGURES Figures 1—2. Distal abdominal structures of Rhagoletis pomonella. 1, Segments 4—8 and genitalia, ventral view. 2, Postabdominal sterna and syntergosterna, ventral view; arrow indicates point of attachment to hypandrium ............................... 129 Figure 3. Genitalia of Rhagoletis pomonella. right lateral view .................................. 130 Figures 4—5. Genitalia of Rhagoletis pomonella. 4, Ventral view. 5, Left oblique view ................................................................................................................................ 131 Figure 6. Genitalia of Rhagoletis pomonella. Longitudinal section, left lateral View ................................................................................................................................ 1 3 2 Figures 7—8. Epandrium and associated structures of Epochra canadensis. 7, Left lateral view. 8, Anterior view (proctiger omitted). Arrows indicate external sulcus (Figure 7) and internal apodeme (Figure 8) ...................................... 133 Figures 9—10. Epandrium and associated structures of Oedicarena Iatifrons. 9, Left lateral view. 10, Anterior view ............................................................................ 134 Figures 11—12. Epandrium and associated structures. '11, Oedicarena Iatifrons, posterior view. 12, Paraterellia immaculate, left lateral view ................................. 135 Figures 13—15. Epandrium and associated structures. 13, Carpomya schineri and 14, Rhagoletis cerasi, left lateral view. 15, Hhagoletis cerasi, right surstylus, medial view .................................................................................................. 136 Figures 16—18. Epandrium and associated structures of FIhago/etis berberidis. 16, Left lateral view. 17, Tip of left surstylus, lateral view. 18, Anterior view ......................................................................................................................................... 137 Figures 19—20. Epandrium and associated structures of Rhagoletis cingulata. 19, Left lateral view. 20, Anterior view (proctiger and setae on right surstylus omitted) .......................................................................................................................... 138 Figure 21. Epandrium and associated structures of Hhagoletis magniterebra, left lateral view .................................................................................................................... 139 Figure 22. Epandrium and associated structures of Rhagoletis magniterebra, anterior view (proctiger omitted) ................................................................................ 140 Figures 23—24. Epandrium and associated structures of Rhagoletis psalida. 23, Left lateral view. 24, AnteriOr view (proctiger omitted) ........................................... 141 Figures 25—26. Epandrium and associated structures of Fihagoletis striate/Ia. 25, Left lateral view. 26, Anterior view (proctiger omitted) .................................... 142 Figures 27—28. Epandrium and associated structures of Trypeta inaequalis. 27, Left lateral view. 28, Anterior view ............................................................................ 143 Figures 29—30. Epandrium and associated structures of Zonosemata electa. 29, Left lateral view. 30, Anterior view ............................................................................ 144 Figures 31—32. 31, Tips of surstyli, Acidia cognata, ventral view. 32, Micrograph, surstyli, Rhagoletis pomonella, posterior view ...................................... 145 Figures 33—38. 33—34, Left bacilliform sclerite, Rhagoletis pomonella. 33, Lateral view. 34, Medial view. 35—36, Prensisetae, posterior view. 35, Rhagoletis altemata. 36, Acidia cognata. 37, Left half epandrium and surstylus, Flhagoletis pomonella, medial view. 38, Micrograph, genitalia, Rhagoletis pomonella, posterior view ............................................................................................. 14 6 Figures 39—41, Genital structures and proctiger. 39, Bacilliform sclerites (diagrammatic), Rhagoletis pomonella, posterolateral view. 40, Proctiger (slide-mounted), Rhagoletis pomonella, ventral view. 41, Phallus, Cryptodacus tau, left lateral view ...................................................................................................... 147 Figures 42—45, Distiphallus. 42—43, Paraterellia immaculate, right and left lateral views. 44—45, Oedicarena persuasa, right and left lateral views .................. 148 Figures 46—49, Distiphallus, right and left lateral views. 46—47, Trypeta inaequalis. 48—49, Epochra canadensis ....................................................................... 14 9 Figures 50—55, Distiphallus, right and left lateral views. 50—51, Rhagoletis adusta. 52—53, Rhagoletis cerasi. 54—55, Rhagoletis cingulata .............................. 1 5 0 Figures 56—61, Distiphallus. 56—57, Hhagoletis nova, right and left lateral views. 58—59, Rhagoletis pomonella, right and left lateral views. 60, Rhagoletis pomonella, dorsoapical view. 61, Flhagoletis magniterebra, dorsolateral view ........................................................................................................... 151 Figures 62—67, Distiphallus, right and left lateral views. 62—63, Rhagoletis psalida. 64—65, Rhagoletis striatella. 66—67, Rhagoletis suavis ............................ 152 Figures 68—69, Distiphallus, right lateral view. 68, Chetostoma rubidium. 69, Micrograph, Rhagoletis suavis ............................................................................... 153 Figures 70—71, Micrographs, distiphallus, right lateral view. 70, Rhagoletis pomonella. 71, Rhagoletis suavis ................................................................................. 154 Figures 72—73, Phallus. 72, Rhagoletis comp/eta, apical view. 73, Phallus ground plan. right lateral view (cross sections: bold lines = sclerotized, plain lines = membranous) ..................................................................................................... 155 Figures 74—75. 74, Generalized wing showing venation and names of wing bands. 75, Evolution of banded wing patterns in the Trypetini. (a) Hypothetical ground plan pattern. (b—d) Evolution of wing pattern with proximal subcostal xi band prominent and band r-m complete. (e—g) Evolution of wing pattern with distal subcostal band prominent and band r-m truncated. See text ............................. 1 5 6 Figures 76—85. Wing patterns of trypetines. 76, Epochra canadensis. 77, Chetostome curvinerve. 78, Euleie fretrie. 79, Zonosemata electe. 80, Pereterellie ypsilon. 81, Oedicarena nigre. 82, Rhegoletis pomonelle. 83, Rhagoletis zoqui. 84, Rhegoletis chionenthi (normal wing shape). 85, Rhagoletis chionenthi (abnormal wing shape). Scale bars equal 1mm ........................ 157 Figures 86—89. Wing patterns of Rhegoletis. 86, Rhegoletis feuste (normal wing shape). 87, Hhegoletis feusta (abnormal wing shape). 88, Rhegoletis jug/endis (normal wing shape). 89, Rhegoletis jug/endis (abnormal wing shape). Scale bars equal 1mm ...................................................................................... 158 Figure 90. Transformation series for wing patterns in Rhegoletis. (a) Rhegoletis blanchardi, (b) Rhegoletis striate/le, (c) Rhagoletis cerasi, (d) Rhegoletis complete, (e) Hhegoletis indifferens, (f) Rhegoletis cingulete, (g) Rhegoletis tebellerie, (h) Rhegoletis zephyrie. Circles on vein R4+5 indicate position of campaniform sensilla. Scale bars equal 1mm. Drawings of wings were made using a drawing tube attached to a stereo microscope ................................. 15 9 Figure 91. Relationship between condition of apical band and ratio of distance between the two distal sensilla on vein R4+5 (distance A) to the distance between the distal most sensillum and apex of vein R4+5 (distance B) in R. cingulete and R. indifferens ................................................................................................................. 1 6 0 Figures 92—93. Scanning electron micrographs of setae of Rhegoletis species. 92, Right orbital setae of R. pomonelle showing longitudinal grooves. 93, Right upper frontal seta of R. complete showing oblique striations lying in longitudinal grooves ........................................................................................................................... 1 61 Figures 94—95. Scanning electron micrographs of scutal microtrichia of Rhegoletis pomonelle. 94, Microtrichia from lateral microtrichiose stripe. 95, Microtrichia from between sublateral and lateral microtrichiose stripes .................. 162 Figure 96. Mean ratio of distance, measured from transverse suture, of the supra-alar seta (spal s) to the dorsocentral seta (dc 3). Number in parentheses after species is sample size; range of the ratio is given above bars. See text ............. 1 63 Figure 97. Mean position of crossvein r-m along vein M. Number in parentheses after species name is sample size; range of the ratio is given above bars. See text ................................................................................................................. 1 64 Figure 98. Mean number of setae on vein R4+5 dorsally beyond branching of vein Rs in Rhegoletis. Number in parentheses after species is sample size; range in number of setae is given above bars .......................................................................... 1 65 Figure 99. Mean number of setae on vein R4+5 dorsally beyond branching of vein Rs. Numbers in parentheses after species is sample size; range for number of setae is given above bats ............................................................................................ 16 6 Figure 100. Transformation series for medial and sublateral pattern systems of tergal coloration. See text ............................................................................................. 167 xii Fig. 101. Percent of species by symmetry system within subfamilies. Number in parentheses after subfamily is sample size. See text .............................................. 1 6 8 Figures 102—103. Scanning electron micrographs of denticles on eversible ovipositor sheath of Rhagoletis species. 102, R. carnivore. 103, R. pomonelle ........ 169 Figure 104. Strict consensus tree of 18,691 cladograms of length = 135, CI = 0.417, and RI = 0.748 (uninformative characters ignored). Bars represent synapomorphies; numbers refer to characters in Table 7. Subtribes are indicated along the top of the tree ............................................................................ 170 Figure 105. Strict consensus tree of 1,000 cladograms of length = 135, CI = 0.430, and RI = 0.848. Bars represent synapomorphies; numbers refer to characters. Subtribes are indicated along the top of the tree ....................................... 171 Figure 106. Strict consensus tree of 12,061 cladograms of length = 37,433, CI = 0.671, and RI = 0.900 (reweighted data). Bars represent synapomorphies; numbers refer to characters in Table 7. Subtribes are indicated along the top of the tree ........................................................................................................................... 1 72 Figure 107. Strict consensus tree of 13,100 cladograms of length = 306, CI = 0.281, and RI = 0.789. Bars represent synapomorphies; numbers refer to characters in Table 7. Subtribes are indicated along the top of the tree ..................... 1 73 Figure 108. Number of polymorphic species by character ......................................... 174 xiii INTRODUCTION The genus Rhegoletis includes 62 described species occurring in temperate areas of the Holarctic, Oriental, and Neotropical regions (Bush, 1966; Hardy, 1977; Foote, 1981, 1984; Berlocher, 1984; Hernandez-Ortiz, 1985, 1993; Norrbom, 1989). The genus is placed in the subtribe Carpomyina (tribe Trypetini) (Foote et el., 1993) which also includes the Palearctic genera Carpomye, Goniglossum, and Myioperdelis; the Nearctic genus Zonosemete; and the Neotropical genera Ctyptodacus (=Lezce), Heywardine (=Cryptoplegie), Rhegoletotrypete, end Stoneole (Norrbom, 1989). Members of Carpomyina whose biology. is known breed in the fleshy fruits of plants from a wide variety of families (see Foote, 1981; White and EIson-Harris, 1992; Hernandez-Ortiz, 1993; Norrbom, 1994; Smith and Bush, in review). A number of these files are serious agricultural pests, especially species of Rhegoletis (Boiler and Prokopy, 1977; Foote, 1981; White endElson-Harris, 1992). Species of Rhegoletis also have been the subject of numerous studies in the field of evolutionary biology (Feder et el., 1988; Bierbaum and Bush, 1990; Frey and Bush, 1990; Bush 1992; Berlocher et el., 1993; Johnson et el., 1996; McPheron and Han, submitted; Smith and Bush, in review). Comparative studies of fruit flies in general, and Rhegoletis in particular, are hindered by the current state of the classification of the family. Recent classifications of the Tephritidae (e.g., Hardy, 1973; Foote et el., 1993) have changed little since the one proposed by Herring in 1947 (Hardy 1980, 1983; Hancock, 1986; Foote et el., 1993). These classifications are untested, intuition-based arrangements, and the degree to which they reflect phylogenetic relationships is uncertain. 1 2 Phylogenetic systematics, or cladistics, is currently the most widely accepted method for inferring phylogenetic relationships (Forey et el., 1992; Kluge and Wolf, 1993). Recent studies have done little to improve the classification of the family, and most suffer from what Kluge and Wolf (1993) called “ad hoc methods that only beer the label, not the meaning, of phylogenetic systematics.“ In Hancock's (1986) classification of the Trypetinae, characters were polarized using an 'outgroup comparison with other subfamilies' without specifying which subfamilies. Further, many of his defining characters are polymorphic (e.g., “Female typically with three spermatheca, two in a few species and genera, and a variously shaped aculeus [segment 8];"), tautological (e.g., 'Leg with a row of bristles on fore femora present or absent"), or noncharacters (e.g., '...; stigma [wing cell sc] not vestigial;...‘). Hancock stated that, “Character trends therefore need to be applied if a workable classification is to be achieved, accepting that various anomalies may occur." However, trends are highly subjective and dividing them into meaningful characters can be quite arbitrary. Discussing the classification of North American fruit flies, Foote et al. (1993) concluded that "homoplasy (convergent evolution) appears to be common in many morphological characters that have been the main basis of classification." Their conclusion was based on the assumption that the family is 'a relatively recent, rapidly radiating group“ (Foote et el., 1993). However, demonstrating homoplasy depends on a resolved cladogram because homoplasy is a property of characters only within the framework of ancestor-descendant relationships (Wiley, 1981). Similarly, Foote et al.'s (1993) use of 'monophyly," 'synapomorphy,“ and 'pleisiomorphy' is often inappropriate because these terms are relative only in conjunction with a testable hypothesis of relationships (cladogram). Another problem has been the assumption that widely distributed characters are primitive. In a phylogenetic study of selected tephritid flies using ribosomal DNA, Han 3 and McPheron (1994) stated that, “When two equally parsimonious interpretations of ancestral states were possible..., the state more common within the Tephritidae was arbitrarily assigned as ancestral" (reference to specific characters omitted). However, when relationships are not resolved, the assumption that common equals primitive does not ensure that the most recent common ancestor to the study group had the primitive state, especially in groups where homoplasy is common (Wiley, 1981; Watrous and Wheeler, 1981). In addition, the cladogram upon which they based their outgroup relationships (Han and McPheron, 1994, figure 1) misrepresented the phylogeny proposed for the Tephritoidea by McAlpine (1989). McAlpine (1989, figure 116.3) places the Piophilidae in a clade that is a sister taxon to the clade containing the Tephritidae, not basal to the Tephritidae as shown by Han and McPheron (1994, figure 1). What should be apparent from the above discussion is the central role that characters play in reconstructing phylogenetic relationships (see also Neff, 1986; Pimentel and Riggins, 1987; Bryant, 1989). During a phylogenetic analysis, it is only in the character analysis that hypotheses can be proposed and tested by deduction (Bryant, 1989). No matter what cladogram we generate, it can, in principle, be explained by induction. Because we can never know when we hit upon the true phylogeny, one scenario is, in principle, as good as another. The confidence that we can have in any phylogeny depends directly upon the characters used to infer it. The work reported herein attempts to make character analysis the central issue. Male genital characters are used extensively in fruit fly taxonomy, but much remains unknown about their structure and homologies. Chapter 1 deals with the morphology of the genitalia of male trypetines in anticipation of their use in phylogenetic analysis. Wing patterns also provide important characters, and, like male genitalia, much of what is known about them is based on taxonomic utility rather than sound morphological study. To stimulate interest in the historical development of wing patterns, and to 4 stabilize the nomenclature of pattern elements, a heuristic model of trypetine wing pattern evolution is presented in Chapter 2. The results of a phylogenetic analysis of relationships among Rhegoletis and related genera are reported in Chapter 3. The phylogenetic analysis consists of two parts: an extensive qualitative analysis of morphology, and a cladistic analysis based on the resulting characters. Throughout this dissertation the terms “figure," ”table," and “character“ are used to refer to the figures, tables, and characters of other authors while "Figure," ”Table,“ and 'Character" refers to those herein. CHAPTER 1 MALE GENITALIA IN THE TRYPETINI (DIPTERA: TEPHRITIDAE) "...men makes nothing so complex as an ant or a fruit fly, and if he did, it 'would surely be subject to errors of construction and assembly..." — Garcia-Bellido et al. (1979) Characters of the male genitalia are commonly used in the taxonomy of fruit flies, and much of what we know about genital morphology is a result of taxonomic studies (e.g., Benjamin, 1934; Aczél, 1955; Bush, 1966; Hardy, 1973; Novak, 1974; Stoltzfus, 1977; Fteidberg and Mathis, 1986; Korneyev, 1986; Norrbom et al., 1988; Stoltzfus, 1988; White, 1988; Hernandez-Ortiz, 1993; Merz, 1994). . In particular, Munro (1947) summarized terminology up to 1947 and gave an extensive description - of tephritid genitalia based on a revision of African species. More recently, Munro (1984) gave a detailed account of genitalia in his revision of dacine fruit flies. Despite this long-standing familiarity with the genitalia of male tephritid flies, there are surprising gaps in our knowledge of the structures. This is in part because descriptions are often based on taxonomic convenience rather than well-reasoned morphological study. As a result, terms'are applied as a matter of personal preference . or taxonomic tradition, and there is often more than one term for a given structure or the same name is given to structures that are not homologous. Another barrier to understanding tephritid genitalia has been disagreement over interpretation of homologies in the male genitalia of the Diptera (summarized by Cumming et al. [1995]). In a recent series of papers (Wood, 1992; Sinclair et al., 1994; Cumming et al., 1995), however, competing hypotheses were evaluated and new homologies proposed. This important body of work codifies terminology and uses 5 6 phylogenetic analysis to corroborate homologies to a greater extent than previous studies (e.g., Griffiths 1972; McAlpine, 1981a, 1989; Wiegmann et al., 1993). Interest in cladistic analysis of tephritid taxa (e.g., Berlocher, 1981; Han, 1992; Norrbom, 1993, 1994; Han et al., 1993) accentuates the need for phylogenetically informative characters. Male genitalia is a potentially rich source of characters for phylogenetic studies, however, absence of uniform terminology and established homologies presently precludes many comparisons. Therefore, following Wood (1992), Sinclair et al. (1994), and Cumming et al. (1995), I present a comprehensive description of male trypetine genitalia; discuss homologies within the family; and propose a ground plan for the phallus of tephritid flies. Materials and Methods Speciesand number of specimens examined are listed in Table 1. Specimens used for dissection were relaxed in a humidor overnight. About two-thirds of the abdomen was excised and macerated in sodium hydroxide (ca. 10%) heated to 60° C until structures cleared (ca. 20—90 min). Abdomens were then acidified in glacial acetic acid for at least 30 min, rinsed in distilled water and stored in micro vials containing glycerin; microvials were attached to the pin below the fly. Stereo and phase-contrast microscopes were used to examine genitalia. Glycerin was used to make temporary microscope slide mounts. Drawings were made using a drawing tube attached to the microscope. When available, frozen or recently killed flies were used for preparations studied with scanning electron microscopy. Specimens were cleaned by soaking in enzymatic laundry detergent (Procter and Gambel's ERA®, 5% v/v) for 30 min with brief (10 sec.) sonication followed by three rinses in double distilled water. Flies were fixed in FAA (2 formalin:1 glacial acetic acidz10 80% EtOH:7 water) for 12—24 h, rinsed three times in 70% EtOH with a 15 min soak between rinses, and dehydrated in a graded 7 alcohol series. Flies were either air dried or dried in a critical point drier, then coated with gold and examined in a JEOL JSM-3SCF scanning electron microscope at the Center for Electron Optics, Michigan State University. Terminology follows Wood (1991), Sinclair et al. (1994) and Cumming et al. (1995) unless noted otherwise. For the purpose of discussion, orientation of the phallus is fully extended posteriorly. Description Genitalia was examined from 278 specimens in 90‘species (Table 1). The following description is based on these examinations. Segments 1—5 (Preebdomen). Terga 1 and 2 are fused and form syntergum 1+2 (tergites 1 and 2). The remaining preabdominal sclerites are free. The pleura are usually unmodified, but an invaginated sac-like structure occurs in the pleural membrane between segments 4 and 5 in Myoleje limate. Segments 6—8 (Postabdomen). Syntergostemum 6+7 is formed by fusion of segments 6 and 7 on the left side of the abdomen (Figure 1—2). Fusion of segments 7 and 8 form a lobe, syntergosternum 7+8, that is continuous with syntergosternum 6+7 (Figures 1—2). Sterne 6 and 7 are narrow and free medially and broad and free on the right. Sternum 7 sometimes has a sharp bend near its middle. The right end of sternum 7 is narrowly attached to the anteromedial edge of the hypandrium (Figure 2, arrow). Stema 6 and 7 each have a pair of sensory setulae. These sensilla are named here according to the sternum and side on which they occur. For example, sensilla GR and 7L occur on the right and left sides, respectively, of sterna 6 and 7 (Figure 2). A small blister- or sac-like structure of unknown, function sometimes occurs medially in the membrane between sterna 6 and 7 (Figure 1). This structure varies from a low swelling that is just detectable to a conspicuous lobe. The structure is evidently an evagination of the intersegmental membrane; it is most easily seen in 8 abdomens treated with NaOH prior to removing genitalia. The membrane between sterna 6 and 7 is taut and flat in specimens without the structure. ~ Epandrium. Tergum 9, the epandrium, is convex dorsally and bears a pair of ' posteroventrally directed surstyli ventrally (Figures 1, 3—6). Posterior to the epandrium is the anus-bearing segment, the proctiger (Figures 1, 3—6). The epandrium is closed ventrally by the subepandrial membrane which runs anteriorly from the proctiger to the base of the phallus (Figure 6). The epandrium bears a number of macrotrichia and often sparse to dense microtrichia. Surstyli. Surstyli vary in length (Figures 7—31) and have the outer surface relatively flat to strongly convex. Apically, surstyli vary from more or less blunt to sharply pointed. Each surstylus bears one to three lobes. The anterior surstylar lobe (Figure 31) occurs on the anteromedial surface of the surstylus near the level of the prensisetae (e.g., Figures 8, 15, 22, 28, 30); it is identified by numerous denticles and one or more (rarely none) sensory setulae on its posterior surface (Figure 32). The posterior surstylar lobe (e.g., Figures 10, 15, 21, 26, 28) may be a small posteroapical lobe (e.g., Figure 31) or a major portion of the entire surstylus (e.g., Figure 21). It bears a number of setae that are often larger than the sensory setulae on the anterior lobe (Figure 32); in Rhegoletis berberidis, there also are a number of small peg-like sensilla distally (Figure 17). Denticles sometimes occur on the posterior surstylar lobe, where they may be confluent with those on the anterior lobe (Figure 15). The medial surstylar lobe (Figures 28, 31), when present, is between the anterior and posterior lobes and is similar in size to the anterior lobe. The medial lobe also bears denticles. Bacilliform sclerites. A bacilliform sclerite is closely associated with the inner surface of each surstylus (Figure 6). In lateral view (Figures 6, 33—34), the bacilliform sclerite is a more or less rod-shaped structure with a twist (often indicated by a notch or groove) usually anterior to midlength and with a pair of apical or 9 subapical prensisetae. The inner and outer prensisetae usually occur at about the same level, but in some species one or the other may be more distal (Figures 35—36). The portion of the bacilliform sclerite posterior to the twist lies in a depression on the inner surface of the surstylus (Figure 37). The bacilliform sclerite and surstylus are connected across the depression by the subepandrial membrane, which usually bears microtrichia (Figure 37). The dorsal edge of the bacilliform sclerite is fused for a variable distance along the posteromedial edge of the surstylus (Figures 32, 38). A dorsal keel that is often erose or serrate sometimes occurs distally on the bacilliform sclerite (Figures 16, 19). A sclerotized bridge between the posterior portion of the right and left bacilliform sclerites is often present (Figures 14, 39). This posterior bridge is formed by an arm running dorsomedially from each bacilliform sclerite to the subepandrial sclerite (Figures 6, 14). In Epochra-canadensis, a membranous connection extends from the bridge to an internal sclerotized process at the base of the surstylus (Figure 8, arrow); externally the process is indicated by a sulcus (Figures 7, arrow). Another sclerotized bridge always occurs anteriorly between left and right bacilliform sclerites (Figures 14, 39). The bridge was fractured medially in a number of specimens (e.g., Figure 22) suggesting the presence of a suture or area of weakness. The subepandrial membrane runs from the proctiger and base of the surstyli to the posterior bridge, and from there to the anterior bridge and phallus (Figures 6, 39). The portion of the subepandrial membrane running from the bridge to the anterodorsal base of the phallus nearly always bears denticles (Figure 6). The anterior end of the bacilliform sclerite usually forms a lobe that projects forward beyond the anterior bridge for a short distance (e.g., Figure 16, “bacilliform sclerite“); this lobe is absent in the Oediacerene, Pereterellie (Figures 9, 12), and Streuzie species. The lobe usually has fibers attached to it from muscles removed during dissection. A muscle runs obliquely forward from the ventral surface of the bacilliform 1O sclerite to the lateral wall of the epandrium. Subependriel sclerite. The subepandrial sclerite (e.g., Figures 6—10, 40) lies within the epandrium above and between the bacilliform sclerites and usually just ahead of the hypoproct. The sclerite is usually small, but in P. immaculate it is quite large (Figure 12). A muscle runs laterally or dorsolaterally from each side of the subepandrial sclerite (Figure 40) to the inner surface of the epandrium; these muscles help identify the sclerite. Phallus. The intromittent organ, the phallus, arises anteromedially to the epandrium (Figures 3—6). When at rest, most of the phallus is concealed beneath tergum 5 (Figure 38) in a pouch that is formed by the intersegmental membrane. A small portion of the phallus is normally visible to the right of the epandrium where it is held against the abdomen by the pregonite (Figure 38). The phallus can be divided into a proximal basiphallus and distal distiphallus (McAlpine, 1981a) (Figure 3—5, 41). The basiphallus and distiphallus can usually be distinguished: they join at an angle, the phallus narrows at their junction (Munro, 1947), and the ventral sclerotized strips of the basiphallus terminate at its apex. The parameral sheath forms the external wall of the phallus and encloses the aedeagus; it has both sclerotized and membranous components. The aedeagus is continuous with the sperm sac via the ejaculatory duct (Figures 3, 6). The aedeagus is membranous for the length of the basiphallus and upon entering the distiphallus, but it often terminates in a sclerotized acrophallus (Figure 41). The basiphallus (Figure 3—5, 41) is relatively long and narrow, and coiled or convoluted. It usually has numerous transverse grooves dorsally and a pair of sclerotized strips ventrally that run its length and which may be fused proximally (Figure 41). A narrow ring-shaped sclerite encircles the base of the basiphallus (Figures 6, 41). The ejaculatory duct and accessory gland enter the basiphallus through the center of the sclerite (Figure 6). The sclerite articulates ventrally with the 11 phallapodeme and is connected to the subepandrial membrane dorsally (Figure 6). In several species, the basiphallus bears a small dorsal or dorsolateral bladder-like structure distally, here termed the basiphallic vesica (Figure 42). A small, irregularly-shaped sclerite sometimes occurs near the apex of the basiphallus (Figures 42—43, 45—46). A pair of membranous ventral keels (Figure 41) occur on the basiphallus in several species. P. immaculate has a pair of small, sclerotized tubercles on the ventral surface of the basiphallus proximally. The distiphallus (Figures 3—5, 41—70) usually is distinctly swollen and much shorter than the basiphallus. The apex of the aedeagus is usually enclosed by the distiphallus, but in E. canadensis (Figures 48—49), 0. Iatifrons, and P. superbe, the aedeagus terminates externally. There is an appressed flap laterally (“ventral flap" of Munro, 1984) formed by a longitudinal invagination of the parameral sheath (e.g., Figures 42, 45, 48, 69). This appressed flap is part of a sclerotized plate that makes up a variable portion of the external wall of the distiphallus. The flap wraps around the distiphallus and the position of its distal edge varies from right lateral to dorsal (c.f. Figures 42 and 45). The distal edge of the flap usually can be identified by microtrichia along its length (e.g., Figures 42, 45, 62, 69). The microtrichia may run the entire length of the flap (Figure 56) or be limited to it's edge near the base of the distiphallus (Figure 68). A membranous flap of variable size occurs apically (Figures 42—70). This apical flap may be cleft (Figure 70), and in most species examined it is microtrichiose (Figure 69) or has fine striations on its internal surface; in some species (e.g., R. suavis group), the flap is also arenose. A variously shaped lobe occurs subapically on the left side of the distiphallus (odd numbered Figures 43—59, 63—67). This subapical lobe varies considerably among the species examined. In some it appears as a simple membranous lobe (Figures 51, 61, 65), while in others it bears additional lobes (Figure 52), microtrichia (Figures 56— 1 2 60, 62—63. 66—67, 69—71), or sclerotized denticles or hooks (Figures 42 inset, 68). Sclerotized plates or strips are sometimes present in the wall of the lobe (Figure 48—49, 57, 62—63), and a lumen could be seen within the lobe in a number of specimens. In the Zonosemata species, the subapical lobe is continuous with the apical flap and does not form a separate lobe as in the other species examined. Structures that appeared to be campaniform sensilla (Figure 64) occurred in a few specimens and in no specific location on the parameral sheath. Some specimens of E. canadensis (Figure 49), O. letifrons, and Trypeta ineequalis (Figure 46) had a minute setiform sensillum ventrobasally on the sheath. Internal structure of the distiphallus is complex and affinities are uncertain. In nearly all specimens examined, the course of the aedeagus through the distiphallus could not be traced (see also Munro, 1947, p. 78). The apex of the aedeagus forms sclerotized tubes or strips in species where it terminates beyond the parameral sheath (Figures 48—49). When the aedeagus is enclosed, its apex often forms a sclerotized acrophallus that resembles either a corrugated plate (Figure 72) or two to three troughs (Figures 50, 62). In several species, however, the apex of the aedeagus could not be discerned. A small, sclerotized loop (= 'valve' of Munro, 1984 and [7] 'basalring' of Merz, 1994) within the base of the distiphallus on the left side occurs in number species (e.g., Figure 63). The sclerotized plate within the parameral sheath usually has at least a small . amount of weak striate, crenulate or rugose sculpturing (inset, Figures 54—56,-58— 59, 62—67). In some species, however, the sculpturing is quite extensive and much more elaborate, forming distinctive polygons, striations and denticles (inset, Figures 42—44, 46—50, 68). A serrate sclerite (Korneyev, 1986) occurs in the distiphallus of S. intermedia and S. Iongipennis (c.f. Stoltzfus, 1988, figure 36). Sperm pump. The ejaculatory duct runs from the sperm sec to the ring-shaped sclerite at the base of the phallus (Figures 3, 6). Entering the phallus with the 13 ejaculatory duct is an elongate accessory gland (Figures 6, 41); the gland is delicate and easily damaged during dissection. A large, spatulate apodeme, the ejaculatory apodeme, attaches to the sperm sac anteriorly (Figures 3—5). The distal edge is usually thin and coplanar with the blade (Figure 3), but it is flattened perpendicular to the blade in a few species. Hypendrium. Sternum 9, the hypandrium, is a simple, U-shaped sclerite that articulates with the anterior edge of the epandrium (Figure 4—5). A laterally compressed, anteriorly projecting lobe, the hypandrial apodeme, occurs medially in several species. A small piece of sternum 7 sometimes remains attached to the hypandrium after dissection and may be mistaken for the hypandrial apodeme in species where the apodeme is absent. Specimens of some species (e.g., Fl. cingulete and R. suavis groups) sometimes have the anterior edge of the hypandrium more or less U-shaped in the transverse plane. An invaginated sac ('genital ring membrane pouch“ of Bush, 1966, figure 75) sometimes occurs in the hypandrial membrane anteriorly. This hypandrial sac varies from relatively shallow and ill defined (Figure 6) to deep and decidedly sac-like; it is lined with numerous, well-sclerotized denticles in the R. suavis species group (Bush, 1966, figures 72—73, 75—76, 78). Pregonites. A small rod-shaped pregonite occurs proximally on both sides of the hypandrium (Figures 3—5, 38). The rods articulate with the medial bases of the hypandrium and the lateral arms of the phallapodeme. Both pregonites are deflected ventrally, but the right one usually more so. When the phallus is not in use, its base is held against the abdomen by a small lobe formed'by the membrane that runs between the right pregonite, hypandrium, and right lateral arm of the phallapodeme (Figures 3, 38) . Phallapodeme. The phallapodeme (aedeagal guide + aedeagal apodeme of McAlpine, 1981a) is a more or less cruciform sclerite (horizontal plane) occurring in the hypandrial membrane (Figures 4—6). It articulates posteriorly with the ring-shaped 14 sclerite at the base of the phallus and laterally with the pregonites. The more or less spatulate anteromedial projection of the apodeme (aedeagal apodeme of McAlpine, 1981a) serves for muscle attachment. Proctiger. The terminal abdominal segment, the proctiger (Figures 1, 3, 6), bears the anus apically and the hypoproct ventrally. The proctiger varies from relatively short (e.g., Figure 14) to relatively long (e.g., Figure 13). In most species studied, the hypoproct forms a somewhat ill-defined sclerite running the length of the proctiger (e.g., Figures 7, 16, 30, 40). The hypoproct typically becomes wider distally and varies in length from about as long as wide to decidedly longer. Apically, the sclerite may be rounded, truncate or bilobed. The hypoproct is divided medially for most or all of its length in the Oedicarena and Pereterellie species (Figures 10—11). In some species, (e.g., Acidia cognate, Chetostoma spp., Streuzie spp.) the hypoproct extends dorsally and forms lateral plates that cover much of the proctiger (Figures 27, 31). The rectal lining within the proctiger of several species has numerous bumps and folds; the rectal lining is smooth in most species. Discussion Segments 1—5. The function of the invaginated pleural sacs in male My. limate are unknown. Male Anestrephe suspena extend pleural pouches during courtship (Nation, 1972); glands associated with the pouches appear to be a source of male sex pheromone (Nation, 1981). Pleural glands, known or suspected to produce courtship odors, occur in a number of male tephritid flies (Jenkins, 1990 and references therein). Segments 6—8. A-sharp bend near the middle of sternum 7 occurred in some or all specimens of about half of the species of Hhagoletis, but only four of the 39 non- Rhegoletis species examined. The sensory setulae on sterna 6 and 7 are found throughout the Diptera, usually occurring on the anterior margin of the sclerites where they serve as landmarks of segmentation (Griffiths, 1972; McAlpine, 1981a). In species studied 1 5 here, some specimens had one or more sensilla in the intersegmental membrane behind the sterna. Sensillum 7R was the most likely to occur in the membrane. In some specimens, one or more sensillae were missing. Surstyli. The term surstylus is used here for the structure referred to in the recent literature (e.g., McAlpine, 1981a) as the outer surstylus. Surstyli are secondarily derived articulated clasping structures in the Eremoneura (Cumming et al., 1995). The cyclorrhaphan surstylus is a lateral outgrowth of tergum 9, the outer surface of which is formed by the epandrium and the inner surface by the bacilliform sclerite (Cumming et al., 1995). Surstyli are not articulated in the Tephritidae, but published illustrations indicate that articulation occurs in several sister groups (e.g., Steyskal, 1958, figures 8, 10, 20, 22 [Richardiidae]; McAlpine, 1987, figures 14— 15 [Lonchaeidae]; Steyskal, 1987a, figure 9 [Platystomatidae]; Steyskal, 1987b, figure 7 [Pyrgotidae]; McAlpine, 1981b, figures 3—4 [Pallopteridae]). Non- articulated surstyli could be a result of the fusion of articulated lobes, or the secondary gain of non-articulated lobes. The sulcus at the base of the surstylus of E. canadensis (Figure 7, arrow) may represent a line of fusion between the epandrium and a once articulated surstylus. Further, the membrane running from the internal apodeme formed by the sulcus to the posterior bridge between the bacilliform sclerites may represent a primitive connection between the surstylus and the subepandrial membrane. However, more study, including detailed evaluation of musculature, is needed to determine which course of evolution occurred in the Tephritidae. Homology of surstylar lobes has not been well established. Benjamin (1934, figure 6) described the anterior surstylar lobe as “a pad (a small soft structure, presumably sensory)." Stoltzfus (1977) used the "dorsal lobe' (=posterior surstylar lobe?) extensively in his taxonomic study of Eutrete. Munro (1984) described anterior and . posterior surstylar lobes and detailed the variation in the lobes of dacines. Munro (1984) identified the anterior lobe by presence of a 'papillose patch or rugose area,“ 16 but did not mention whether sensilla also are present. Norrbom et al. (1988) noted that the 'Fihegoletis type“ surstylus is longer than the 'inner surstylus' (=bacilliform sclerite) and has the 'mesal" (=anterior) lobe near the level of the prensisetae; parenthetically, they state that the 'Fihegoletis" type surstylus is secondarily absent in Zonosemeta. Anterior and posterior lobes were present in most species examined here. In the Zonosemata species (Figure 30) and E. canadensis (Figure 8) only the anterior lobe was present. All three lobes were present only in A. cognate and T. ineequelis (Figure 28, 31 ). An anterior surstylar lobe was present in all species examined. Its homology is established by, its position relative to the prensisetae and by the presence of denticles and usually one or more sensillae. From specimens studied, Munro's (1984) work cited above, and illustrations in the literature (e.g., Munro, 1947; Bush, 1966; Drew, 1972; Hardy, 1973; Stoltzfus, 1977; Freidberg, 1980; Stoltzfus, 1988; Jenkins and Turner, 1989; Korneyev, 1991; Condon and Norrbom, 1994; Norrbom, 1994), it is likely that an anterior lobe is a feature of surstyli throughout the Tephritidae. Function of the anterior lobe is unknown, but its close apposition to the prensisetae suggest that it is used in clasping. The medial and posterior surstylar lobes are more problematical. When both lobes are present, they can be identified simply by their positions. In most species studied, however, only one lobe in addition to the anterior lobe is present. Whether this lobe is derived from the medial or posterior lobes, or both, could not be determined. However, development of the posteroapical portion of the surstylus in a number of species (e.g., Figures 11, 15, 23—24) suggests derivation from the posterior lobe. Loss or reduction of medial and posterior lobes may result in surstyli like those found in Zonosemeta (Figures 29—30) and numerous tephritines (e.g., Jenkins and Turner, 1989; Novak, 1974; Stoltzfus, 1977). 17 Bacilliform sclerites. The structures referred to here as bacilliform sclerites are the inner surstyli of McAlpine (1981a). Munro (1947, 1984) termed the structures “twisted rods“ in reference to their characteristic twist. The bacilliform sclerites are formed by secondary sclerotization of the subepandrial membrane (Sinclair et al., 1994; Cumming et al. 1995). It is very likely that the bridges and denticles described. for the species studied here are also derived from the subepandrial membrane by secondary sclerotization. The anterior bridge appears to be widespread within the Tephritidae (see Bush, 1966; Drew, 1969; Stoltzfus, 1977; Freidberg and Mathis, 1986; Condon and Norrbom, 1994; Norrbom, 1994). Following Griffiths (1972), Korneyev (1985) and Norrbom and Kim (1988) termed the anterior bridge “interparameral sclerite.“ However, Griffiths' (1972) term is inappropriate if the structure is derived from the subepandrial membrane, which is likely (see Cumming et al., 1995). The posterior bridge in the species studied ranged from well-developed, sclerotized structures to a simple membranous connection between the bacilliform» subepandrial sclerites. Phallus. There has been much confusion over the naming and homology of structures. of the phallus. The structure herein termed basiphallic vesica has been described as a “gland-like tubular sac“ near the apex of the basiphallus (Bush, 1966); a “membranous bladder“ at the base of the distiphallus (Drew, 1969); a “basal gland“ of the distiphallus (Munro, 1984); a “fold or ligule“ at the “place of articulation“ of the distiphallus with the basiphallus (Korneyev, 1985); a “membranous lateral lobe“ at the base of the distiphallus (Norrbom et al., 1988); and a “basal lobe“ of the distiphallus (Condon and Norrbom, 1994). Neither Bush (1966) nor Munro (1984) stated why the structure should be considered glandular. Eberhard (1990, figure 6) showed that the structure is expandable (like a vesica) in Ceretitis capitete . Confusion exists over which part of the phallus the vesica belongs to because the extent of the distiphallus and basiphallus has not been clearly defined. Distinction between basiphallus and 18 distiphallus is usually apparent when using the limits listed above. The terms distiphallus and glans are often used synonymously, but the latter is more appropriate for the terminal portion of the vertebrate penis. Similarly, the term prepuce and its derivatives (see Korneyev, 1986) are more appropriate to vertebrate morphology. The structure described herein as the subapical distiphallic lobe is widely distributed and has been referred to as “apical appendage“ (Bush, 1966), “apical process“, (Foote, 1981), “apicodorsal rod“ (Munro, 1984; Han, 1992), “juxta“ (Korneyev, 1986; Merz, 1994), “accessory sclerite“ (White, 1988), “tubular structure“ (Eberhard, 1990), and “apical spinose appendage“ (Hernandez-Ortiz, 1993; Condon and Norrbom, 1994). In all specimens examined, the base of the lobe is subapical, although the apex may, extend beyond the tip of the distiphallus. Except for Nearctic Chetostoma, the apex of the subapical distiphallic lobe in the non-carpomyines examined is trumpet-shaped (but it is often flattened in preparations). This shape is similar to the “tubular structure“ of Ce. capiteta (Eberhard 1990, figure 6) and the “apicodorsal rod“ of dacines (Munro, 1984), in which the subapical lobe appears to be well sclerotized. Nearctic Chetostoma have the subapical lobe elongated and with a pair of sclerotized hooks apically. The subapical lobe of the carpomyines studied is usually an attenuated lobe or flattened flap; it is usually bare but is sometimes fimbriate or bears microtrichia that vary in size and density. Position of the lobe on the distiphallus of carpomyines is similar to that of the other species examined. Function of the subapical distiphallic lobe is unknown. Eberhard (1990, figure 6, caption) states that inflation of the apical membranous portion of the distiphallus (“second expandable sac,“ labeled “b“) of Ce. capiteta drives the subapical lobe (“tubular structure,“ labeled “0“) into “a cone in the wall of the vagina“ (=ventral receptacle?). Eberhard (1990) implies that sperm is transferred through the subapical lobe; unfortunately, no reference or further discussion is given. However, because the 19 subapical lobe is an outgrowth of the parameral sheath, and because the aedeagus is often recognizable as a definite acrophallus, it seems more likely that sperm is transferred through the aedeagus. Further, in all material examined the subapical lobe appeared to be closed distally, which substantiates Munro's (1984) observation of a “concave- convex cap at the tip“ of the lobe in dacines. Munro (1947, p. 79) called the “terminal membranous part of the aedeagus [=distiphellus]“ the “vesica.“ This is equivalent to the apical membranous flap described herein and the “second expandable sac“ illustrated by Eberhard (1990). In the species studied here, the apical membranous flap varies from small to large (e.g., Figures 62 and 68). Han's (1992) “dorsal sclerite“ is an area of the sclerotized plate within the parameral sheath of the distiphallus and not an actual sclerite. Like the dorsal sclerite, the “median granulate sclerite“ (Han, 1992) is not a sclerite, but a sclerotized area of the phallotheca that bears denticles or papillae, the extent of which is quite variable. The epiphallic sclerite of Korneyev (1985) is interpreted here as the acrophallus. The term epiphallic sclerite is better applied to the small, bilobed plate that occurs at the extreme base of the phallus in some tephritines (e.g., Tephritis spp.). The phallus of tephritid flies is a potential source of characters for phylogenetic studies. However, little progress will be made towards understanding evolution of the phallus until a ground plan is proposed. I therefore propose the following model. The phallus is in the form of a tube within a tube (Figure 73). The outer tube is derived from the parameral sheath and the inner tube from the aedeagus (Cumming et el., 1995). The phallus can be further divided along its proximal-distal axis into two more or less well defined regions: a proximal basiphallus and a distal distiphallus. The aedeagus is mostly membranous, but may terminate in a sclerotized acrophallus; it is fused to the parameral sheath within the distiphallus. The parameral sheath consists of membranous and sclerotized elements. A sclerotized plate occurs within the parameral 20 sheath of the distiphallus where it makes up a variable portion of the outer wall and internal structure. Longitudinal infolding of the parameral sheath of the distiphallus produces an appressed lateral flap. This flap wraps around the distiphallus and its extent determines the size of the terminal opening of the parameral sheath. A variously-shaped subapical lobe is formed by an outgrowth of the parameral sheath near the apex of the distiphallus. The wall of the subapical lobe may be completely or partially sclerotized, or entirely membranous. The lobe may bear various superficial processes such as microtrichia, denticles, and supernumerary lobes. The apex of the distiphallus forms a membranous apical flap that may bear microtrichia, denticles or other superficial outgrowths (e.g., arenosity). Identification of these structures or their derivatives will provide characters for phylogenetic analysis. For example, the trumpet-shaped subapical distiphallic lobe of the trypetines studied is likely due to descent rather than chance (see Chapter 3). Moreover, the position and shape of the lobe is very similar to that of the lobe in dacines (c.f. Munro, 1984; Han, 1992). As another example, membranous and sclerotized portions of the parameral sheath of the distiphallus are more or less coextensive in the species studied here. In the putatively derived genus Tephritis, however, the membranous component of the sheath is much larger than the sclerotized portion (Jenkins, 1985; Jenkins and Turner, 1989; Merz, 1994). Evidently, during the evolution of Tephria's, the membranous component of the parameral sheath has become enlarged relative to the sclerotized portion. Ejeculatory apodeme and phallapodeme. Shape“ and size of the ejaculatory apodeme and the anteromedial portion of the phallapodeme (= aedeagal apodeme of McAlpine, 1981a) is age-dependent. These structures continue to grow for a period of time after adult emergence (Pickett, 1937; Drew, 1969; Drew, 1972; Kamali and Schulz, 1974; Berube, 1978; Munro, 1984). With a few exceptions (e.g., R. ribicoIe—see Bush, 1966), use of these structures as taxonomic characters (e.g., Bush, 1966; Novak 2 1 1974; Stoltzfus, 1977; Foote, 1981) should be viewed with skepticism (Drew, 1972; Munro, 1984). Hypendrium. The invaginated sac occurring in the hypandrial membrane of some species has been termed “genital ring membrane pouch“ by Bush (1966) and “membranous process of the hypandrium“ by Korneyev (1986). Munro (1984) reported a “fultella [=hypandrium] gland“ in dacines with free terga and, in dacines with fused terga, the gland is replaced by a sac that is setulose or bare. The terms “hypandrial sec“ or “hypandrial gland“ would be more consistent with current terminology. A hypandrial sac is widespread in the Tephritidae, occurring in Dacines (e.g., Munro, 1984), trypetines (e.g., Bush, 1966), and tephritines (e.g., Korneyev, 1986i Pregonites. The rod-like structure associated with the base of the hypandrium and articulating with the phallapodeme laterally (herein termed pregonite) has been called “inferior rod“ (Munro, 1947), “lateral sclerite of the hypandrium“ (Korneyev, 1986), “intermediate rod“ (Munro, 1984), and “lateral sclerite“ (Han, 1992). Homology of the structure is not clear. In the ground plan of the Diptera, gonocoxites are closely associate with, but separate from, the hypandrium (Wood, 1991). Many Nematocera and some Lower Brachycera retain this primitive condition (Wood, 1991; Sinclair et al., 1994). In many other Lower Brachycera, the gonocoxites are partially to completely fused to the hypandrium (Sinclair et al., 1994). The hypandrium and gonocoxites are completely fused in the Eremoneura, and structures that are secondarily derived from gonopods in the Schizophora are termed pregonites (Cumming et al., 1995). The position of the rod-like structures at the base of the hypandrium in tephritid flies suggests a gonopodal origin. Articulation of the rods with the lateral arms of the phallapodeme, which is itself secondarily derived from the gonocoxal portion of the hypandrium (Cumming et al., 1995), indicates that the rods are secondarily derived. The term pregonite should therefore be used for these rod-like structure. 22 Proctiger. The portion of the rectum lying within the proctiger is convoluted or papillate in the Chetostoma, Eulie, Myoleje, and Streuzie species. These convolutions and papillae are similar to those seen in a specimen of Ce. capitata examined. Rectal glands in male Ce. capitata are the source of a sex pheromone (Nation, 1981) that is produced during courtship and is highly attractive to females (Prokopy and Hendrichs, 1979i Although based on relatively few trypetines, most of the results of this study should be applicable to the entire family. In order for knowledge of genital morphology to expand, a unified system of terminology must be settled upon, and refinement of homologies sought. As more is learned about the structure and function of tephritid genitalia, previously untested characters can be incorporated into phylogenetic studies. CHAPTER 2 A HEURISTIC MODEL OF WING PATTERN EVOLUTION IN THE TRYPETINI (DIPTERA: TEPHRITIDAE) “The primary value of models is heuristic. ...the establishment that a model accurately represents the 'ectuel processes occurring in a reel system“ is not even a theoretical possibility.“ —Oreskes et al. (1994) The wings of tephritid flies often bear color patterns. Wing patterns may consist of dark bands on a hyaline field, hyaline spots on a dark field, or a combination of bands and spots. The relative ease with which wing patterns are observed has long made them a useful character in tephritid taxonomy (Cole, 1969). Patterns are often characteristic of a species and many flies can be identified on the basis of wing pattern alone. Patterns also are useful for identifying seasonal color morphs within species (Jenkins and Turner, 1989), and for distinguishing some genera and tribes. Although wing patterns are very useful for identification, a system based on taxonomic utility presents challenges to systematists wanting to use wing patterns for phylogenetic analysis. Phylogenetic relationships usually are overlooked by taxonomists interested in finding taxonomic differences rather than characters that unite taxa (i.e., synapomorphies). Unfortunately, taxonomic differences are often homoplasious or autapomorphic. One problem in using a taxonomically based system for phylogenetic research is that the names of pattern elements are inconsistently applied (Table 2), sometimes even within a single work (e.g., White and Elson-Harris, 1992, figures 37— 38, 97; Foote et al., 1993, pp. 129, 248, 325). Taxonomists also devise systems of wing pattern nomenclature for particular groups (e.g., White and Elson-Harris, 1992, figures 35—36; Condon and Norrbom, 1994), but such esoterica make comparing patterns among groups with different systems uncertain at best. 2 3 24 Another important consideration for phylogeneticists is a general lack of demonstrated (or even proposed) homology of wing pattern elements. In species with banded wing patterns, bands are named based on their relative position, and as a result, bands with the same name may not be homologous. ln Foote et al. (1993), for example, the subbasal and discal bands of Epochra canadensis, Chetostoma califomicum, and Ch. rubidium are the discal and intercalary bands, respectively, of Rhegoletis. A heuristic model of the evolution of banded wing patterns like those found in the Trypetini is presented below. The purpose for presenting such a model is to stabilize nomenclature for banded wing patterns and to provide a hypothetical basis for constructing transformation series of pattern evolution for phylogenetic analysis. A discussion of possible mechanisms involved in pattern formation follows the model. Evolution of Banded Wing Patterns in the Trypetini As it becomes evident, that between these different types of design a genetic connection really exists, so that they can be arranged in a series, leading from the most primitive and regular to the farthest modified and most capricious, and that this series is the same for different interrelated genera and families, the conclusion, that this correspondence roots in relationship, is a natural one. — van Bemmelen (1917) The fundamental difference in wing patterns is the extent and position of pigmented and hyaline areas of the wing membrane. Different wing patterns evolve by the expansion and contraction of these areas (Aciurina provides a compelling example of this [see Foote et al., 1993, figures 112—124]). For simplicity, only the expansion of hyaline areas will be considered in the hypothetical model presented here. Although the model is described in terms of the expansion of hyaline areas, it should be borne in mind that it is the pigmented portions of a pattern that determine the extent of the hyaline areas. To avoid comparing non-homologous bands, names used here (Figure 74), except for the apical band, are based on morphological landmarks instead of their relative position on the wing. Figure 75a gives a hypothetical wing pattern from which other 25 patterns are derived. Choice of the positions of hyaline areas for the hypothetical pattern was based on requirements of the model and actual patterns with extensively pigmented areas and relatively few small hyaline marks (e.g., Aciurina spp. [Steyskal, 1984], Xenthaciura spp. [Aczél, 1950, 1952], Acanthonevrini [Hardy, 1973, 1974, 1986], and African spp. [Munro, 1947]). Relatively few hyaline areas are needed to derive patterns of the species studied. New patterns are the result of the inward expansion of marginal hyaline areas and the enlargement of discal hyaline spots. Patterns illustrated in. Figure 75 show some of the changes in the ground plan that are needed to obtain banded wing patterns. The patterns were not taken from particular species, but elements of each can be found in real patterns in the taxonomic literature. Band h (humeral bend). Band h runs posteriorly from the costa at or near the level of vein h usually to the level of vein CuA2; vein h is its landmark (Figure 74). Band h is sometimes indistinct, especially in species where the proximal portions of the wing are extensively pigmented (e.g., Euleia spp. [Figure 78], Myoleje spp., Streuzie spp.), or the pattern is generally lightly pigmented (e.g., Cerpomya incomplete, R. jug/andis [Figures 88—891). The humeral band is formed by the inward expansion of hyaline spots in cells c and bc and the anal lobe and alula (Figure 75). Additional hyaline spots in the extreme wing base (e.g., the base of cell br) also may be involved. In most carpomyines studied, band h is free, crosses cells bm and cup, and covers vein CuA2 (e.g., Figures 82—87). In E. canadensis (Figure 76), Ch. californicum, and Ch. rubidium, band h runs to or across the base of cells bm and cup and does not cross vein Cqu. The humeral band is truncated posteriorly or interrupted by a hyaline area or spot in cell brn in a number of non- carpomyines. In several of these (e.g., Acidia cognate, Ch. curvinerve [Figure 77], Oedicarena latifrons, Pereterellie ypsilon [Figure 80]), the posterior portion of both band h and the proximal subcostal band (described below) converge on vein Cqu. It is unclear whether the posterior portion of band h actually belongs to that band or to the 26 posterior portion of the proximal subcostal band (see below). Bend sc (subcostal band). The subcostal band runs posteriorly from cell 30 (its landmark) usually to the wing margin or nearly so (Figure 74). The band may be entire or divided into proximal and distal bands. Each band may be partially fused to other wing bands (e.g., Figures 78—79). The proximal edge of the subcostal band is formed concomitantly with the distal edge of the humeral band. Coalescence of a hyaline spot in cell br with a hyaline area in cell cua1 divides the band into proximal and distal portions (Figure 75). The distal edge of band sc is formed by the inward expansion of hyaline areas in cell r1 and cua1 and their coalescence with spots in cells br or dm. When the proximal hyaline area in r1 and the hyaline area in cua1 converge on and coalesce with the spot in br, the distal portion of band sc is reduced or obliterated forming a prominent proximal band sc (Figure 75b— d). The proximal subcostal band runs posteriorly from cell sc usually to the level of vein Cqu (Figures 76—77). When the hyaline areas in r1 and cua1 converge on and coalesce with the spot in dm, both the proximal and distal portions of band so are prominent (Figure 7Se—f). Subsequent loss of the proximal portion results in a prominent distal band sc (Figure 759). The distal band runs posteriorly from cell sc to cell cua1 or the posterior wing margin, and at least its distal edge crosses vein r-m (Figures 79—89). Both subcostal bands are prominent in Ch. curvinerve (Figure 77), but in most species seen only one band is prominent although a second faint or incomplete band can sometimes be traced (e.g., Figures 76, 79—80). Norrbom et al. (1988) suggested that the pigmented spots lying on vein h and cell “bcu' (=cell cup) in Oedicarena (e.g., Figure 81) may be an “incomplete subbasal band“ =humeral band) and, as such, a possible synapomorphy. However, as interpreted here, the spot on the humeral crossvein is part of the humeral band and the spot on cell cup is either part of the humeral band or proximal band 30; similar spots occur in Pereterellie. Foote et al. (1993) did not recognize a proximal subcostal band and as a 27 result their “discal“ band is actually the distal subcostal band in Rhagoletis and band r- m (described below) in Epochra and Chetostoma. Norrbom (1993) also identified band r-m in E. canadensis and E. mexicana as the “discal“ band. It is important to keep in mind the distinctions between proximal and distal subcostal bands in phylogenetic studies because they are not strictly homologous. Band r-m (radial-medial band). Band r-m runs posteriorly from the costa in cell r1 across vein r-m, its landmark (Figures 74, 76, 80).' In a number of species, however, the band extends posteriorly only to vein R4+5 (e.g., Figures 79, 81) or is absent (e.g., Figures 82—89). Band r-m is formed by the inward expansion of hyaline areas in cells r1 and cua1 and their coalescence with spots in cells br, dm, or both (Figure 75). If the proximal hyaline area in cell r1 coalesces with the spot in cell br, band r-m crosses vein r-m (Figure 75b—d). If the hyaline areas in cell r1 converge on and coalesce with the spot in cell dm, band r-m is truncated and does not cross vein r-m (Figure 75e—f).' In both cases, the distal edge of band r-m is formed by the coalescence of the distal hyaline area in cell r1 with the hyaline areas in cells dm and cua1. When band r-m crosses vein r- m it either runs as a free bend to the posterior wing margin (or nearly so) (e.g., Figure 76), or it joins distal band sc (e.g., Figures 78, 80). When the band is truncated, it extends posteriorly only to vein R4+5 (e.g., Figures 79, 81). Band r-m joins the apical band anteriorly in several species (e.g., P. varipennis, P. immaculate, P. superba, Euleia spp. [Figure 78] and specimens of E. canadensis). lt- joins the apical bend anteriorly and band dm-cu posteriorly in P. superba and the Euleia species (Figure 78). Band r-m is absent in a number of species, especially North American FIhago/etis (e.g., Figures 82—89), and some specimens of Rhagoletotrypeta rohweri and Rh. uniformis. Norrbom (1989) reported that the truncated band r-m is rare in the Tephritidae and that it may be a synapomorphy for a taxon that includes the Carpomyina and 28 Oedicarena. Foote et al. (1993) did not recognize the homology of a band crossing vein r-m with the truncated band r-m (their “intercalary“ band; Table 2). Norrbom (1994) suggested that the “apical fork“ (=band r-m) in the “discal“ band (=band so + band r-m) of P. ypsilon (Figure 80) may be homologous to the “accessory“ band (=truncated band r-m) of some carpomyines, but did not recognize this homology in other Pereterellie species. Band dm-cu (discal medial-cubital band). Band dm-cu runs from the posterior wing margin across vein dm-cu, its landmark, and usually joins the apical band anteriorly (Figure 74). The proximal edge of band dm-cu is formed concomitantly with the distal edge of band r-m. Expansion of a hyaline area in cell m forms the distal edge of band dm-cu (Figure 75). Anteriorly, band dm-cu usually is continuous with the apical band (Figures 77, 79—80, 83—87, 89). In a number of species, band dm-cu may join band r-m, band sc (e.g., Figure 79), or both (e.g., Figure 78). Band dm-cu is incomplete in several species (e.g., Figure 81). Because band dm-cu and the apical band are usually joined anteriorly, it is unclear whether they evolved as a single pattern element or separately. However, some evolutionary independence is needed to explain differences (e.g., reduction) observed in apical bands without complementary changes in band dm-cu. Foote (1981) identified the subcostal band as the “preapical' band (=band dm-cu, Table 2) and band dm-cu as the posterior apical band in the Ft. pomonelle species group. Further, Foote et al. (1993) did not recognize the homology of band dm-cu (their “subapical“ band) in the Ft. pomonelle group with this band in other Rhagoletis species. They considered the “subapical“ band to be absent in the pomonelle group, and the band crossing vein dm-cu to be a posterior apical band (Foote et al., 1993). This is despite having defined their “subapical“ band as the band crossing vein dm-cu (Foote et al., 1993, p. 325), which is clearly crossed by a band in the pomonelle group. In none of 29 the species studied here is a posterior apical band present and bend dm-cu simultaneously absent. Apical band. The apical band, unlike the other bands, has no structural landmark. Distally, the band ends in the wing margin beyond the wing apex; anteriorly, it usually joins band dm-cu (Figure 74). The apical band may occur as a definite band occupying much of the wing apex (e.g., Figure 76), or as small, variously shaped marks (e.g., Figure 81). The band may be continuous with the wing margin (e.g., Figure 79) or separated from it by a narrow hyaline area (Figure 82). It may be entire (e.g., Figures 79—80) or divided into anterior and posterior bands (e.g., Figures 78, 84—87). The posterior edge of the apical band is formed concomitantly with the distal edge of band dm-cu by the inward expansion of a hyaline area in cell m (Figure 75). Inward expansion of a second hyaline area in cell m divides the band into anterior and posterior apical bands. In all species studied except C. incomplete, the apex of the wing is at least partially pigmented; in C. incomplete, the wing apex is hyaline. The apical band in E. canadensis, P. varipennis, P. immaculate, P. superbe, and the Euleia species joins band r-m (at least narrowly) or is continuous with an area of pigment along the costa. It joins bends dm-cu and so anteriorly in the R. pomonelle group (Figure 82) and Ft. zernyi. Assuming that a divided apical band joined to band dm-cu anteriorly and continuous with the wing margin is the pleisiomorphic condition (as in Figure 75d, f—g), the following changes may occur to produce apical bands like those observed in this study. A pleisiomorphic apical band has the distal comer of the posterior band ending well behind vein M and the distal corner of the anterior band ending at or near the apex of M (e.g., Figure 78). Loss of the posterior band results in an apical band that is undivided and continuous with the wing margin (e.g., Figures 77, 79). Several species (e.g., R. nova and R. psalida groups, R. magniterebre and Eu. uncinata) have the posterior apical band reduced to varying degrees. Secondary division of the remaining (anterior) apical band 30 may result in an apical band like that in the R. cingulete species group (Figure 84). In these species, the distal corner of the new posterior band ends in the wing margin at or near vein M and the distal corner of the new anterior band (or spot) ends at the margin well ahead of M. It is important to keep in mind that anterior and posterior apical bands that are the result of secondary division are not homologous to the pleisiomorphic apical bands. Loss of the anterior band in a cingulete-like pattern would result in a hyaline area between the posterior band and wing margin (Figure 82). The resulting apical band would be like that of C. schineri, C. vesuviene, Goniglossum wiedemenni, Myioperdelis perdalina, and several Rhagoletis species. Bush (1966) suggested that the wing pattern of Ft. ribicola, which has an undivided apical band separated from the wing margin by a hyaline area, could be derived by loss of the anterior apical band from a cingulete-like pattern. Bush (1966) also suggested that the pattern of Ft. berberis could be derived by loss of the posterior apical band of a cingulete-like pattern. However, the single, undivided apical band of R. berberis is continuous with the wing margin and its distal comer is at or near the apex of vein M. It derived from a cingulete-like pattern, the distal corner of the band would be well ahead of vein M and at least a small hyaline area would lie between the band and wing margin in cell r1 and r3...4. A simpler explanation of the pattern of R. berberis is that the posterior band has been lost from a pleisiotypic apical band as described above. Foote et al. (1993) refer to any single undivided apical band as the anterior apical band, whether it is continuous with the costa or separated by a hyaline area, and without regard to precisely where it ends in the wing margin. The posterior apical band is defined by Foote et al. (1993) as originating on either the “subapical“ band (=band dm- cu, Table 2) or the “discal“ band (=distal subcostal band) and ending on or between the tip of veins CuA1 and M. They consider the apical band of the R. cingulete group to be divided, presumably secondarily so. 'The apical band of some species of walnut-infesting FIhagoletis (Fl. boycei, R. 31 juglendis, R. remosee, Ft. zoqur) provides yet another modification. In these flies, streaks of pigment lay along or between veins R4+5 and M (Figure 83). Because apical bands typically cross the radial and medial veins obliquely, these streaks are likely novel marks rather than a result of the reduction of pre-existing bands. Figure 90 provides an example of a transformation series showing changes in wing patterns of Rhegoletis. The pleisiotypic pattern (Figure 90a) has all wing bands observed in the genus and is represented by species like those in the ferruginea species group. Loss of band r-m results in the pattern found in the striatella group (Figure 90b), while loss of the posterior apical band results in patterns like that in several Eurasian species (e.g., R. berberidis, R. cerasi [Figure 900], R. caucasica). Loss of both bands (Figure 90d) is seen in species such as those in the suavis group, R. berberis, Fl. emiliee, R. flevicincta, and R. reducta. Secondary division of the apical band results in a cingulete group pattern (Figure 90e—f). Loss of the anterior arm of the apical band in the cingulete group produces patterns like those in the label/aria group (Figure 909), Ft. batave, R. flavigenuelis, R. mango/ice, and R. ribicole. Fusion of the distal three bands in the anterior half of the wing produces the pomonelle group pattern (Figure 90h). Within the series additional modifications may alter patterns. For example, fusion of the humeral and subcostal bands posteriorly in the pomonelle group (Figures 82, 90h) and some tabellerie-like patterns (Figure 909). It is one thing to arrange wing patterns into plausible transformation series, but it is quite another to assert that such evolution has occurred in nature. After all, hypothesis testing, not judging plausibility, is the task of science. In order to test hypotheses of character evolution systematists require hypotheses of phylogeny. It is the interplay between these two types of hypotheses that determines the veracity of each. It now appears that Rhegoletis is not monophyletic, and not all monophyletic groups in this assemblage have been identified (McPheron and Han, submitted; Smith and Bush, in review; Chapter 3 herein). However, portions of the above transformation series 32 pertinent to established monophyletic groups could be used to test relationships within and between those groups. Mechanisms of Wing Pattern Formation The mechanisms of wing pattern formation in tephritid flies is unknown, but some inferences may be made from observations of adult wings. In order to discuss pattern formation, however, it is necessary to have in mind some general features of wing development and models for pattern formation in animals. Therefore, a brief summary of each is given below. Because details of wing development in tephritid flies sufficient for studying pattern formation haVe not been reported, the wing development of Drosophila melanogester has been summarized; the summary is based on the works of Weddington (1941), Bainbridge and Bownes (1981) and Johnson and Milner (1987). Wings develop from imaginal discs. Many of the structures of the wing, including dorsal and ventral surfaces, sensilla, basal sclerites, and some veins, are determined in the imaginal disc (Campuzano and Modolell, 1992, figure 1). The dorsal and ventral surfaces of the wing each develop as a two-dimensional cellular monolayer of epidermis. Basement membranes of the wing surfaces are fused except where blood lacunae form. Lecunae run longitudinally through the disc and around its margin, and provide the only means by which material enters and leaves the developing wing. Tracheae invade lacunae during the prepupal period and form the primary venation of the wing. Evagination of the disc occurs during the prepupal stage. Shortly after evagination and onset of the pupal stage, the wing epidermis undergoes a period of rapid growth by cell division. Near the middle of the pupal period, secondary tracheae that will form the adult wing veins replace the primary tracheae. A second period of growth occurs at about the middle of the pupal stage and as a result the wing becomes pleated and folded upon itself. The wing expands by enlargement of epidermal cells during this second period of growth. Chitin is deposited during the last half of the pupal period. After emergence, wing 33 epidermis degenerates leaving only the nonliving cuticle, and the wing expands to its adult form. The pupal period of D. melanogester lasts about 93—105 h at 26° C. Current models of pattern formation propose that cells destined to produce integumental pigment are determined by a prepattern formed early in development (Bard, 1977; Murray, 1981, 1988; Nijhout, 1985, 1991). Prepatterns are the result of reaction-diffusion systems that generate stable patterns of activators and inhibitors of varying concentrations in the developing integument (Murray, 1981, 1988; Meinhardt, 1982; Nijhout, 1985, 1991; Pool, 1991; see also Lengyel and Epstein, 1991). Timing and the geometry and scale of the integument where chemical interactions occur strongly influence prepatterns (Murray, 1981, 1988). Once established, prepatterns may be modified by allometric growth of the integument; however, the characteristic pattern of a species is to a large extent determined by the prepattern (Bard, 1977; Murray, 1981, 1988; Nijhout, 1991). Patterns become visible when pigment is produced in cells determined by the prepattern. The amount of pigment produced by cells, and therefore the intensity of pigmentation, is determined by the interaction of pigment-inducing morphogens and the prepattern. Two observations suggest that wing patterns in tephritid flies are determined early in development. First, flies with an abnormal wing shape have the same pattern as flies with the normal wing shape (Figures 84—89). In D. melanogester, shape of the adult wing emerges after the wing has undergone its final period of growth (45—60 h postpupariation) (Waddlngton, 1941). If the prepattern is established after wing shape is attained, then differences in shape should affect wing pattern because even small changes in the geometry of a developmental field can alter the prepattern (and thence the final pattern) (Murray, 1981, 1988). Second, wing pattern and distribution of the three distal campaniform sensilla on vein R4+5 dorsally (Figure 90) are strongly correlated. In Rhegoletis.? the distal two sensilla are situated very close to one another in species with band r-m present or with 34 a pleisiotypic (as described above) apical band, or both (Figure 90a—c). Species without band r-m or with a derived apical band, or both, have the middle sensillum decidedly proximal to the distal sensillum (Figure 90d—g), and in the R. pomonelle group, the proximal two sensilla are situated very close to one another (Figure 90h). This correlation may be the result of pattern and sensilla simultaneously tracking wing growth. If so, then pattern and sensilla may be established at about the same time. Development of these sensilla has not been documented for tephritid flies, however, in D. melanogester, precursor cells of the sensilla are established in the wing bud by 12 h after pupariation (Murray et al., 1984; Palka et al., 1986), which is well before expansion of the wing at around 45—60 h. Also, position of the sensilla relative to one another does not appear to be affected by wing growth after precursor cells are established (Murray et al., 1984, figures 1—2; Palka et al., 1986, figures 1—2). These observations conform to the expectation of current models (Bard, 1977; Murray, 1981, 1988; Nijhout, 1985, 1991) that patterns are established early in ontogeny. In lepidopterans, insects for which wing pattern has been most studied, pattern determination begins in the imaginal wing discs during the last larval instar (Nijhout, 1985, 1991). It is reasonable then to suspect that wing patterns in tephritid flies are determined early in wing development, perhaps in the imaginal disc. Allometric growth may affect the arrangement of elements in some wing patterns of tephritid flies. The closely related Fl. pomonelle and Ft. tabelleria species groups (Berlocher and Bush, 1982; McPheron and Han, submitted; Smith and Bush, in review) have wing patterns that are essentially the same except for fusion of the three distal bands (30, dm-cu, apical) in the anterior half of the wing in the pomonelle group (Figures 82, 90h). Fusion of these bands may be the result of retarded growth in that area of the Wing. Spacing of the campaniform sensilla on vein R4+5 (Figure 909—h) and orientation of vein dm-cu also suggests differential growth rates occur. The proximal displacement of the anterior portion of band dm-cu in Euleia species (e.g., 35 Figure 78) may be another example of allometric growth affecting arrangement of pattern elements. Examples of the effect of allometry on the shape of pattern elements may be the step-like distal edge of the apical band in the Ft. pomonelle group (Figure 82). Another . example may be the relationship between the condition of the anterior apical band and spacing of distal sensilla on vein R4+5 in the R. cingulete species group (Figure 90e— f). The anterior apical band is usually broken in R. cingulete but complete in Ft. indifferens, R. chionanthi (Figure 84), and H. osmanthi (Bush, 1966, table 8; Foote et al., 1993). The ratio of the distance between the distal two sensilla to the distance between the distal most sensillum and apex of vein R4+5 of 16 flies (8 6 6 and 8 9 9) each of R. cingulete and R.-indifferens was very significantly different (arcsine [(A/B) - 1] transformation; d.f. = 1, F = 22.87; p = 0.00004) (Microsoft EXCEL, 1992— 1993, single factor ANOVA). Condition of the apical band is not determined solely by allometric proportions, however, as 4 specimens of cingulete had the band complete and 2 of the indifferens had the band broken (Figure 91). Wing patterns of insects have been studied in greatest detail for Lepidoptera, and the excellent work of Nijhout (e.g., 1985, 1991) and his coworkers provides a valuable paradigm for the study of wing patterns infruit flies. There are, however, distinct differences in the wing patterns of lepidopterans and tephritid flies. A fundamental difference is wing morphology and location of pigment. Pigment of the wings of lepidopterans is found exclusively in the wing scales—modified macrochaetae covering the external surface of the wing (Nijhout, 1985, 1991). Macrochaetae are absent from the wings of tephritid flies except as setae or campaniform sensilla on some anterior wing veins; the wing membrane is bare or covered with microtrichia—acellular superficial outgrowths of the integument (McAlpine, 1981a). Microtrichia may contribute somewhat to the color- pattern, as for example the whitish apical spot of euphrantines and the white apical crescent of Eutreta species (see Foote et al., 1993), 36 or as general infuscation of the wing membrane (e.g., Ft. alternate species group). Microtrichia may also produce structural colors. Munro (1947) described “shining silvery spots or areas“ that he termed “argents“ on the wings of a number of African tephritid flies. He reported that argents are caused by “greatly attenuated and colourless“ microtrichia. Nevertheless, the vast majority of color making up tephritid patterns lies within the wing, not in surface structures as in Lepidoptera. In some species, it appears that pigment is laminated between hyaline upper and lower wing surfaces, rather than the cuticle itself being pigmented. This is especially so for the proximal streaks and spots in the wings of species of Ceratitis. Because pigment is a product of the epidermal cells of a wing, the laminated appearance may be explained as pigment left between hyaline cuticle after epidermal cells degenerate. Debris from epidermal cells has been reported between the wing surfaces in Drosophila (Johnson and Milner, 1987, figure 4f). Another essential difference between Lepidoptera and tephritid flies is the form of the patterns themselves. In lepidopterans, two distinct systems combine to form wing patterns: a system of discrete pattern elements is superimposed on a second system that forms a background pattern (Nijhout, 1991). Pattern elements develop along the midline of wing cells (in the venetional sense) and veins act as boundaries to the elements (Nijhout, 1985, 1991). Pattern elements in wing cells on either side of vein M3, which approximates the boundary between the anterior and posterior developmental compartments of Drosophila, are often different (Nijhout, 1991). Wing cells are serially homologous with respect to wing pattern, and pattern elements within each cell develop and evolve independently of those in other cells and the background pattern (Nijhout, 1991). Different areas of the background pattern may evolve independently and there is no correspondence between background and wing structures, such as veins (Nijhout, 1991). Also, overall patterns on upper and lower surfaces of a wing are often . different. 37 The patterns of tephritid flies generally are simpler than those of lepidopterans. It appears that the patterns of fruit flies are formed by a single system similar to the background pattern of Lepidoptera. Pigment is deposited on an essentially colorless field without additional elements superimposed on the pigmented areas, and patterns are identical on upper and lower wing surfaces. The roles of wing cells and veins are unclear, but it does not appear that they necessarily influence pattern. An exception may be the truncated form of band r-m which may run to vein R4+5, but was never seen to cross that vein. The wing margin does, however, seem to be involved in organizing patterns. In numerous species there is a series of hyaline spots running around the margin. Fusion and expansion of these spots appear to be responsible for many of the differences observed in patterns. Interestingly, nearly 80 years ago Van Bemmelen (1917) suggested that banded wing patterns in some Diptera may be derived by expansion and coalescence of marginal spots like those found in tephritid flies with irrorate wing patterns. Unlike lepidopterans, the wing patterns of tephritid flies are not visibly disrupted across the anterior-posterior developmental boundary, which lies between veins R4+5 and M in Drosophila (Garcia-Bellido et al., 1979). Also, the wing cells of tephritid flies do not appear to be serially homologous with respect to wing pattern. However, as in Lepidoptera, it appears that portions of tephritid wing patterns evolve independently. In South American Rhegoletis. for example, (see Foote, 1981; Frias et al., 1987) the anterior apical band may be lost to varying degrees without affecting the presence of the posterior apical band and vice versa. The model presented here stabilizes wing band nomenclature, and provides a framework for constructing transformation series of wing patterns for phylogenetic analysis. The model will not fit all tephritid fly wing patterns nor is it intended to do so. It is hoped that interest will be stimulated for systematically sorting out the remarkable variation in fruit fly wing patterns. An essential step in this process will be delineating a ground plan pattern. Van Bemmelen (1917) suspected that a ground plan exists for 38 the color patterns of dipteran wings, but empirical study is needed to evaluate his rather vague conclusions. Information on the development of tephritid wings and wing patterns is also needed; modern molecular techniques will be invaluable in this regard. The coplanar nature of wings and their distinctive landmarks (veins) should facilitate morphometric analysis. The wing patterns of tephritid flies should provide an ideal system for studying pattern formation in animals; perhaps this intriguing problem will not be left to smolder for another 80 years. CHAPTER 3 PHYLOGENEI'IC ANALYSIS OF FIHAGOLETIS AND RELATED GENERA (DIPTERA: TEPHRITIDAE) “...the search for lost things is hindered by routine habits and that is why it is so difficult to find them.“ — Marquez (1991) Hypotheses of phylogenetic relationships among organisms provide the basis for much of comparative biology (Kluge and Wolf, 1993). Phylogenies are especially important to studies of evolution because they provide a historical framework from which to ask questions and direct research (Miles and Dunham, 1993). Active interest in evolutionary studies of Rhagoletis (e.g., Berlocher et al., 1993) underscores the need for a phylogeny of the genus. Norrbom (1989) placed Rhagoletis in his subtribe Carpomyina, stating that the “monophyly of the genus has not been demonstrated and its relationships to other Carpomyina are poorly understood“ (see also Foote et al., 1993; Norrbom, 1994). Phylogenetic analyses of Nearctic Hhagoletis species have been reported by Berlocher (1981) (morphology and allozymes), Berlocher and Bush (1982) (allozymes), Ming (1996) (ribosomal DNA), McPheron and Han (submitted) and Smith and Bush (in review) (mitochondrial DNA). There has been no phylogenetic analysis of the genus on a worldwide basis to date. A problem in constructing a phylogeny for Rhagoletis has been the choice of an outgroup. Because supergeneric classifications of the Tephritidae (e.g., Hering, 1947; Hardy, 1973; Hancock, 1986; Foote et al., 1993) are untested intuition-based hypotheses, little can be said with confidence about the evolution of major groups within the family. Foote et al. (1993) stated that one reason relationships are poorly resolved for higher taxa is that homoplasy is common in the morphological characters used to 39 4o construct classifications. However, their conclusion is largely anecdotal because the critical character and cladistic analyses necessary to establish family-wide homoplasy have not been carried out. The purpose of this study was to 1) conduct a detailed analysis of the morphology of Rhagoletis and related genera; 2) test the monophyly of FIhegoletis; and 3) identify an outgroup for use in subsequent analyses of intrageneric relationships. Materials and Methods “...[systemetists] devote very little effort, in most cases no effort whatever, to the methods by which characters are recognized or defined.“ (emphasis in original) — Neff (1986) Morphological terminology follows that of McAlpine (1981a) unless otherwise noted. The term species is used herein for the nominal taxa normally dealt with by taxonomists and represented by museum specimens; it was from these specimens that morphological data were obtained. I follow the supergeneric classification of Foote et al. (1993), summarized in Table 3. Species examined and their distribution and larval hosts are given in Table 4. Two undescribed species, Fthagoletis “florida“ and Fihagoletis nr. tabelleria (Table 4), also were included. Whenever possible specimens for study were selected from throughout their species“ range. If available, specimens used to study genitalia were in addition to those used for other structures because removing genitalia often destroys characters on other portions of the abdomen (see Chapter 1 for details on preparing genitalia for study). Light (stereo and compound) and scanning electron microscopes were used to examine specimens (see Chapter 1 for details on scanning electron microscopy). Character Analysis. A list of morphological structures was compiled from McAlpine (1981a), and a preliminary list of qualitative attributes was generated by screening . these structures in one to several specimens of each species. Attributes that appeared to vary discretely were retained for further analysis while those that were invariable or 41 appeared to vary continuously were omitted. This initial, cursory survey was necessary because of the large number of species and attributes examined. Variation of each attribute was then partitioned into provisional states, and the attributes were scored for 1—27 specimens of each species (Table 1). Distribution of states within and between species was summarized, and attributes with more than one state for a given species were re-evaluated. Re-evaluation consisted of re-examining the study specimens and, if necessary, redefining the attribute, its states, or both. After re-evaluation, attributes were retained if their states were found to be discrete even though coextensive within a species. These attributes were the characters used in the cladistic analysis. If an attribute could not be objectively parsed into discrete states or was found to vary continuously it. was eliminated from the study. Cladistic Analysis. The data set was analyzed with PAUP 3.1.1 (Swofford, 1993) on a Power Macintosh® 7100/66 personal computer with 10,000K RAM allocated to the software. Redundant taxa (Table 5) and characters occurring in single species (Table 6) were removed to increase search speed (but see Yeates, 1992). Species represented by a single specimen also were excluded because of the large number of genital characters (27 male, 15 female) that would have to be coded as missing. All characters were unordered and only missing characters were coded as "t.“ The effect of polymorphisms was tested by searching on the data set with polymorphic characters included and removed. Multiple searches were performed using starting trees generated with random and simple addition sequences and Tree-Bisection-Reconnection branch rearranging (Maddison, 1991; Maddison et al., 1992). Random addition searches performed 1,000 replicates with no more than 2 or 3 trees saved during each replicate (Maddison et al., 1992). Searches were allowed to run to completion or were aborted when there was insufficient computer memory to store new trees or the search became excessively slow. MacClade 3.0 (Maddison and Maddison, 1992) was subsequently used to trace character 4 2 evolution. The monophyly of Rhagoletis was evaluated by filtering all minimal length trees with a user-defined constraint tree where Rhagoletis was monophyletic. To identify an outgroup for Rhegoletis. 37 species in 16 genera from the Trypetini were included in the analysis (Table 5). Trees were rooted using Epochra canadensis as the outgroup because it is from the Euphrantini (Table 3), a tribe considered to be primitive to the Trypetini (Hering, 1947; Hardy, 1973; Foote et al., 1993). Results and Discussion “A hypothesis, after all, is no better than the evidence that supports it, and hypotheses without evidence are mere wishful thinking.“ — Barber (1994) Character Analysis A total of 101 species and 879 specimens were included in this study (Table 1). One-hundred and sixty-five morphological structures were examined, and from these 534 attributes were screened for use in the character analysis. Two-hundred and forty-seven of the attributes were analyzed for cladistic characters resulting in 91,941 recorded observations. The final data set (Table 5) included 88 species and 77 characters (Table 7), 28 of which were polymorphic for one or more species. Characters not included in the cladistic analysis are listed in Table 8. Thirteen characters were autapomorphous (Table 6). Head (Characters 1—10, Tables 5, 7) Antenna. A number of species have a dorsoapical point on the flagellum (Character 1). Dorsoapical points range from minute (e.g., Euleia fretrie) to relatively large (e.g., FIhago/etis flavigenuelis), and size usually varies within species. Most species of Rhegoletis have at least a small dorsoapical point, but FIhagoletis caucasice, Rhagoletis kurentsovi, and some specimens of several other species lacked a point. A dorsoapical point also occurs in some specimens of Carpomye, Goniglossum, Heywardine, and Myioperdelis. 4 3 Bush (1966) reported that the flagellum usually bears a dorsoapical point in North American Rhegoletis. but that some Palearctic and Neotropical species have the apex rounded. Foote (1981) also noted the tendency for some Latin American Rhagoletis to have the apex rounded. Berlocher (1981) scored only the pointed state for Rhagoletis boycei, Rhegoletis fausta, Rhagoletis pomonella, Rhagoletis ribicola, and Rhegoletis tabelleria, species that I found to be polymorphic for the character. Norrbom et al. (1988) and Norrbom (1989, 1990) considered the dorsoapical point to be a possible synapomorphy for Rhagoletis and related genera. Norrbom (1989) noted that a dorsoapical point occurs in Carpomye, Cryptoplagie (=Cryptodecus), Heywardine, Myioperdelis. Zonosemate, and most Rhegoletis. He also suggested that the point is lost in Goniglossum, but a minute point was seen in four of the five specimens of Goniglossum wiedemenni examined here. Norrbom (1994) scored Heywardine cuculi and Heywardine cuculiformis as having a “distinct dorsoapical pointed lobe“ whereas I scored these species as polymorphic and without a point, respectively. Most species examined have a microtrichiose arista (Character 2), with the microtrichia ranging from very short (e.g., Pereterellie immaculate) to relatively long (e.g., Rhegoletis striatella). Except for Rhegoletis lycopersella, Rhagoletis tometis, and Rhegoletis mecquerti, South American Rhegoletis species had the arista bare. In R. lycoperselle and R. tometis, a few microtrichia occurred in the proximal half or less of the arista. In addition to sparse proximal microtrichia, specimens of R. mecquarti also had 1—6 microtrichia in the distal half. South American Rhagoletis species often have the arista sinuous, especially distally, and shiny in addition to being bare. Facial Ridge. Comparison of the width of the facial ridge to the parafacial (Character 3) is made at the level of the ventral end of the facial suture. A narrow facial ridge commonly occurs in species that also have the facial ridge about as wide as the parafacial. The facial ridge of Euleia species is distinctly wider than the parafacial. The broad facial ridge in Chetostoma curvinerve is probably due to the extreme 44 enlargement of the setae on the parafacial (Character 9). In Chetostoma californicum and Chetostoma rubidium, these setae are not enlarged and the facial ridge is similar in width to the parafacial. Because of this, the broad facial ridge in Ch. curvinerve is not considered to be homologous to the. broad facial ridge in the Euleia species and was scored the same as Ch. californicum and Ch. rubidium. Chaetotaxy. Color of the genal, gular, postocellar and postocular setae (Characters 4—7) is often lighter than the color of other principal head setae in many of the species studied here. Although there is a trend for these setae to be lighter, there is also considerable intraspecific variation. One or more pair of setae may be lighter in a given specimen, and in a few cases, left and right setae vary in color. In the Carpomye species and Myioperdelis pardaline, the upper orbital and inner and outer vertical setae may also be lighter than other principal setae. Within the Tephritidae, color of principal head setae varies from nearly white to black. Color of these setae is often used taxonomically to help separate subfamilies (e.g., Hardy, 1973, 1974; Foote, 1980), genera (e.g., Munro, 1947; Richter, 1970; Foote and Steyskal 1987), and species (e.g., White, 1988; Foote, 1981; Foote et al., 1993).. . Berlocher (1981) used “light“ and “dark“ states for postocellar, postocular, genal and gular setae. (Berlocher listed postocular setae twice [characters 25 and 31]. Based on the distribution of his states and my own observations, it is likely that he mistakenly used “postocular“ for “postocellar“ in character 25.) Like Berlocher (1981), I originally recorded the color of the genal, gular, postocellar and postocular setae as “light“ or “dark.“ Using relative color appeared to be a good strategy because absolute color may vary depending on lighting (see below). However, deciding if setae are “light“ or “dark“ can be quite arbitrary. In specimens of some species (e.g., Rhagoletis alternate), the difference in color was very slight or some setae were intermediate in color. On the other hand, frontal, orbital and vertical setae were always the darkest (except as noted above for Cerpomya spp. and M. pardaline), 45 often black. Therefore, the color of genal, gular, postocellar and postocular setae was based on a comparison to the color of other principal setae. In this way, character states reflect whether colors are the same (concolorous) or not, without regard to the degree to which the colors differ. It is well known among fruit fly taxonomists that the color of setae often changes with viewing angle, thus making determination of setal color imprecise. This is especially so for light colored setae. It is likely that variable setal color is due in part to the surface ultrastructure of setae. Oblique striations lying in longitudinal grooves of the setae (Figure 92—93) may reflect light differentially as the specimen is turned. Surface structure and the quality of the light reflected from these setae are very similar to that of certain scales on the wings of the moth, Diechrysia (=Plusia) balluce (Ghiradella, 1984, figure 6). Ghiradella (1984) reported that in D. balluce “patches of shiny, satiny scales...brighten and darken with movement of light.“ Ghiradella (1984) explained that this brightening and darkening is due to microribs that extend between longitudinal ridges in the scale so that “[t]he scale thus presents to the incoming light a series of parallel rodlets that selectively reflect or scatter light, depending on their orientation.“ Further evidence for a structural role in setal color comes when ‘ flies are examined in fluid preservatives such as ethanol. Under such conditions the color of setae ceases to vary, probably because of the difference in the refractive index of air and ethanol (see Nijhout, 1991, plate 2). Another source of setal color may be from pigment or some other substance in the lumen of setae. When generally lightly pigmented species like Rhagoletis basiola are viewed in fluid, a dark colored substance is sometimes visible in the lumen of the Iargersetae. The number of frontal and orbital setae is used extensively in tephritid fly taxonomy (e.g., Hardy 1973, 1974, 1980, 1986, 1988; Foote, 1980; Foote et al., 1993; Foote and Steyskal, 1987). Bush (1966) considered three frontal and two orbital setae to be diagnostic characters for Rhagoletis, and Norrbom (1994) regarded 46 four frontal setae to be apomorphic for Zonosemeta. For species studied here, the most common number of frontal setae was three, but the number ranged from one to seven per side and overlapped continuously. Scores for this character by Berlocher (1981) and Norrbom (1994) reflect its polymorphic nature. Because of the high level of intraspecific variation, the number of frontal setae was not used in the cladistic analysis. The number“ of orbital setae (Character 8) for most species studied was two. The upper orbital seta is absent in E. canadensis, two of the four specimens of H. cuculi, and females of the Streuzie species. (Contrary to Foote et al. [1993, p. 373], only male Streuzie lack all orbital setae.) Absence of the upper orbital seta was scored as the derived state even though it is absent in the outgroup, E. canadensis. This is because absence of the seta probably represents a true loss and is therefore derived (Pimentel and Riggins, 1987). In a few specimens where the number of orbitals varied from right to left sides, the most common number of setae in the species was used. Setae on the parafacial, gene, or both, of the Chetostoma species were larger or more numerous, or both, than setae in these areas in other species (Character 9). Enlarged frontal setae (Character 10) occurred in only males of the species of Streuzie examined. Thorax (Characters 11—20, Tables 5, 7) Coloration. The scutum proper is uniformly yellowish, brownish to black (most species), or it has a distinct color pattern (e.g., Cerpomye schineri, Streuzie spp., Zonosemeta spp.). Ground color of the scutum (Character 11) usually does not vary within species. In Rhegoletis complete, the ground color is typically yellowish, but very rarely (2/1,062 specimens examined) there are dark brown morphs. Ground color in four of the six specimens of Rhegoletis blancherdi examined is black; in the remaining two it is yellowish. In Euleia hereclei, there are both yellowish and black flies, and the color 4 7 difference is quite dramatic (see Foote, 1959, p. 149; White, 1988). In Oedicarena letifrons, color varies from yellowish to dark brown. Ground color is polymorphic within the Rhegoletis suavis and Rhegoletis ferruginea groups. Bush (1966) considered ground color of the thorax to be “highly variable“ in Rhegoletis, ranging from yellow to black. Norrbom (1994) likewise commented that the thorax color is “highly variable“ in the Tephritidae. The states used here for scutal ground color are essentially the same as those used by Norrbom for thoracic color (1994, character 15), and species common to each study have identical scores. Scutal patterns are often quite distinctive. The Cerpomya-like scutal pattern (Character 12) consists of a pair of dark postpronotal, notopleural, supra-alar, postalar, scutellar and subscutellar maculae, and a single dark interacrostical macula (see White and Elson-Harris, 1992, figures 210, 233). Extent of the maculae varies somewhat and adjacent maculae may be discrete or fused. Maculae are at least partially covered with velvety black microtrichia that may be worn off in some specimens. The interacrostical macula is divided into anterior and posterior portions in M. pardalina and G. wiedemenni; the posterior portion forms a dark spot on the disc of the scutellum. Only the subscutellar marks are present in Cerpomya incomplete. The Cerpomya-like scutal pattern is similar to the scutal pattern of a number of ceretitines (e.g., White and Elson-Harris, 1992, figures 208, 211). A whitish or yellowish scutal mark occurs medially in several species (Character 12). In the Rhegoletotrypeta and Zonosemate species, the mark is a claviform stripe extending anteriorly from the prescutellar area. In Cryptodacus tau, H. cuculi, H. cuculiformis, P. immaculate, Pereterellie varipennis and Pereterellie ypsilon, the mark is a prescutellar spot or blotch. Additional light and dark markings occur in Cr. tau, Rhegoletotrypeta pestranai, and the Heywardine and Zonosemate species. Norrbom (1994) hypothesized that the four carpomine genera with a scutal white spot (i.e., Cryptodacus, Heywardine, Rhegoletotrypeta, and Zonosemate) form a monophyletic 48 group, but results of his analysis indicated that the group is paraphyletic. A few species have dark scutal marks or stripes that are intraspecifically variable. Specimens of the R. ferruginea group have three or five dark scutal stripes extending forward from the prescutellar area: one medially, a pair sublaterally, and in some specimens, a pair laterally. The stripes vary in width and may be fused anteriorly or posteriorly or both. A single specimen of R. pomonelle from Mil Cumbres, Michoacan, Mexico also had scutal stripes. One or more of the following scutal marks are usually present in the Streuzie species: a medial pair of dark maculae extending posteriorly from the pronotum to the level of the postpronotal setae or a little beyond; a pair of sublateral dark maculae extending from the level of the presutural supra-alar seta anteriorly to about the level of the posterior extent of the anterior medial maculae; a pair of sublateral dark suipes extending anteriorly from the level of the intra-alar setae to the transverse suture; and a pair of lateral dark stripes extending anteriorly from the level of the postalar setae to the transverse suture and passing over the postsutural supra-alar setae (see Steyskal, 1986, figure 8; Stoltzfus, 1988, figure 16). All marks were present in the Streuzie intermedie specimens examined. Some specimens of Streuzie Iongipennis had all maculae except the lateral most stripes, while others had only the anterior-most maculae. In Streuzie perfecta, only the anterior-most maculae were present, but in some specimens even these were absent. Specimens of Oedicarena beemeri, 0. letifrons, and Oedicarena nigra have a dark central spot occupying a variable portion of the scutum. The spot is contained within the area circumscribed by the presutural supra-alar and intra-alar setae in the specimen of O. beemeri, while in the other two species the spot occupies essentially the entire scutum. Coloration of other portions of the thorax have been used extensively in taxonomy and to infer relationships and thus warrant discussion here. The postpronotal lobe in 49 most species is lighter (whitish or yellowish) than the ground color of the scutum. In a few species (e.g., Ch. rubidium, Myoleje Iucida, and the Rhegoletis psalida group), the lobe is concolorous with the scutum. Norrbom (1994) used the states “mostly or entirely white“ and “mostly or entirely brown“ for the postpronotal lobe. The scutellum is uniformly pigmented and concolorous with the scutum (e.g., Chetostoma spp.) or uniformly pigmented and lighter (whitish or yellowish) than the scutum (most Rhegoletis spp.), or has a color pattern. Scutellar patterns range from a simple, dark central spot (sometimes vague) in E. canadensis to relatively elaborate patterns with light and dark elements, as in C. schineri. Bush (1966) and Foote (1981) use coloration of the scutellum throughout their taxonomies of Rhegoletis. Norrbom (1994) considered a generally white scutellum to be pleisiomorphic in his analysis of Cryptodacus, Heywardine, and Rhegoletotrypeta. A whitish or yellowish pleural stripe runs from the postpronotal lobe to the wing base in most species. The stripe usually includes the postpronotal lobe, a variable amount (usually 1/5—1/3) of the upper portion of the proepimeron and anepisternum, a small sclerite (=paratergite?) above the anepisternum, and the pleural wing process including the greater ampulla. In Or. tau, the stripe is interrupted by a dark brown wedge-shaped mark extending from the proepimeron and anepisternum, and is continuous with the transverse suture and anepistemal cleft. Bush (1966) considered a pleural stripe as one of several diagnostic characters for Rhegoletis. The pleural stripe in Oedicarena species may be indistinct or absent, especially in persuase and tetanops (Norrbom et al., 1988; Foote et. el., 1993). Color of thoracic pleura is sexually dimorphic in R. complete, Rhegoletis ramosae and Rhegoletis zoqui (see Bush [1966] and Hernandez-Ortiz [1985] for descriptions). A whitish or yellowish band occurs dorsally on the ketepisternum of specimens of C. schineri, Cr. tau, G. wiedemenni, and the Heywardine and Zonosemate species. Presence of the band was ambiguous in C. incomplete, Cerpomya vesuviene, M. pardalina, P. 50 varipennis, S. intermedie, and the Euleia species. Except for Eu. hereclei, a dorsal band is most apparent in species with relatively more pigmentation on the lower portion or disc of the ketepisternum. In lightly pigmented color morphs of Eu. hereclei, a faint band is present, especially anteriorly, while in the melanic forms the band is absent. This character was not used in the cladistic analysis because of the difficulty in scoring it. Presence of a dorsal white area on the katepistemum was included in Norrbom's (1994) analysis of Latin American carpomyines. Although determining the color of the postpronotal lobe scutellum, and pleuron is usually simple in darkly pigmented flies (e.g., most Rhegoletis spp.), the color in lightly pigmented species can be ambiguous. This may be due to little contrast between the color of these areas and ground color of the thorax, preservation artifacts (decomposition of subcuticular structures—see below), or both. For example, the postpronotal lobe, scutellum, and pleural stripe were lighter than the yellowish ground color of live specimens of Eu. fratria and R. basic/a, but there was usually no difference in the color of these areas in pinned specimens. Another factor affecting color is the method of killing specimens. Specimens killed by freezing often have these areas grayish or brownish, whereas specimens killed by chemical agents or preserved in fluid usually retain the natural color of the areas. Coloration of the thorax may be the result of cuticular pigments, subcuticular structures, or both. Brownish to black elements of scutal patterns appear to be due to cuticular pigments, and in some instances may be associated with areas of muscle attachment. For example, the dark medial presutural marks in the Streuzie species are at the approximate position of the anterior insertion of the dorsal longitudinal flight muscles. Sites of muscle attachment may provide convenient landmarks for homologizing maculae. Other dark markings, such as the stripes in the R. ferruginea group, simply appear to be melanized portions of the integument. Color of the whitish or yellowish pattern elements may be due to subcuticular 51 structures seen through the integument. Scutal cuticle is nearly colorless in newly emerged adults of R. pomonella, and whitish membranous structures are clearly visible through the integument of the postpronotal lobe and scutellum and in the area of the pleural stripe. The structures initially look like collapsed sacs, but within about a day they enlarge and become closely appressed to the inner surface of the integument. As the black ground color of the thorax develops, the cuticle of the postpronotal lobe, scutellum and pleural stripe remains nearly colorless. These subcuticular structures form a soft amorphous mass in specimens preserved in FAA. In pinned specimens treated with NaOH, the integument of the postpronotal lobes, scutellum, and pleural stripe rapidly loses its whitish color while the remainder of the thorax remains darkly pigmented. The yellowish dorsal band on the ketepisternum of several species (e.g., C. schineri, H. cuculi, and the Zonosemate spp.) and the yellowish or whitish elements of the more elaborate scutal patterns (e.g., the Cerpomya-like pattern, Cr. tau, and the Zonosemate spp.) also may be due to the visibility of subcuticular structures. Munro (1984) discussed at length yellow areas of the thorax of dacines that he termed “xanthines.“ My observations above are very similar to those reported by Munro (1984). According to Munro, the xanthines of dacines may be discolored by preservation and become similar in color to the adjacent integument. Further, he reported essentially the same results that I obtained for specimens treated with caustic (potash) and specimens preserved in fluid (alcohol). What are the subcuticular structures that lend whitish and yellowish colors to thoracic patterns? Adult Muscamorphe are characterized as having well developed tracheal air sacs in the thorax, head, and abdomen (McAlpine, 1989). In Drosophila, air sacs in newly emerged flies are collapsed, but within 24 h after emergence they expand to occupy a large portion of the body cavity (Wigglesworth, 1963). Within the thorax of Drosophila, air sacs occur in the postpronotal lobe, scutellum, and along the pleuron (Wigglesworth, 1950)—the precise locations where the subcuticular 52 structures in trypetines are visible. The internal surface of air sacs is hydrophobic and the taenidia are often reduced (Nation, 1985). Microscopic examination of the subcuticular structure removed from the scutellum of an anesthetized R. pomonelle showed the structure to be a delicate hydrophobic membrane studded with small granular objects. Circumstantial evidence suggests that the subcuticular structures seen through the integument of tephritid flies are air sacs, but further study is clearly needed. Vestiture. Scutal setulae vary from dark brown or black to yellowish or white. In a number of species that possess microtrichiose stripes, scutal setulae are a mixture of whitish and brownish or black setulae (Character 14), with the whitish ones mostly associated with the microtrichiose stripes. However, in the Zonosemate species (except Zonosemate vidrapennis), and Cr. tau, species without microtrichiose stripes, color of scutal setulae correspond, at least in part, to the yellowish, or brownish to black color of the integument from where they arise. In species with predominantly light colored setulae (e.g., R. suavis and R. tabelleria groups), the peripheral setulae are often darker than the discal ones. In R. pomonelle and Zonosemate vittigera, there were specimens with uniformly dark setulae and specimens with a mixture of light and dark setulae; these were the only instances of intraspecific variation. The precise distribution of light and dark scutal setulae in species with the mixed state suggests that the state is not homologous among species. Bush (1966) and Foote (1981) make extensive use of patterns formed by scutal setulae in the taxonomy of Rhegoletis. The Oedicarena species are peculiar in having bare spots at the inner ends of the transverse suture and base of dorsocentral setae (Character 20). Norrbom et al. (1988) considered this character to be a synapomorphy for Oedicarena. They also reported that, within the Tephritidae, these bare spots are unique toOedicerene. However, Ore/lie falcata has bare spots in the same locations as Oedicarena, and setulae, microtrichia, or both, are reduced or absent in one or both of these locations in other terelliines and ceratitines. 5 3 Setulae on the postpronotal lobe are uniformly colored or a mixture of whitish and brownish or black setae. Both conditions occurred in 17% (17/101) of the species examined. The character was not used in the cladistic analysis because it was sometimes difficult to judge the color of the setulae (see discussion of setal color in section on head characters above). Berlocher (1981, character 44) divided color of postpronotal setulae into only dark and only light states, but I observed mixed setulae in eight of the species common to his and my studies. Microtrichia are distributed over the entire scutum or limited to its periphery (Character 13). When only peripheral microtrichia are present, they are relatively small and difficult to see. When discal microtrichia are present, they, as well as peripheral microtrichia, are easily observed. When viewed from behind at a low angle, scutal microtrichia are uniformly distributed or form stripes. There are five faint stripes in E. canadensis: one medially, a pair sublaterally, and'a pair laterally. In the other species, the medial stripeis absent and only the sublateral and lateral stripes are present. Further, the dorsocentral seta lies within the sublateral stripe in E. canadensis, but in other species the dorsocentral lies between the sublateral and lateral stripes. Stripes may be free or fused anteriorly, posteriorly, or both; sublateral and lateral stripes are separated by only a narrow line in R. striatella. In R. pomonelle, microtrichia from stripes are bent near their base and somewhat dilated, while interstripe microtrichia are straight and more or less evenly tapered (Figures 94—95) Microtrichiose stripes are used extensively in descriptions and identification of Rhegoletis species (e.g., Bush, 1966; Foote, 1981). Berlocher (1981, character 49) referred to scutal microtrichia as being “complete,“ “partial,“ or “absent.“ From the distribution of these states in Berlocher's data matrix, he evidently was referring to the presence or absence of microtrichia, and whether they are uniformly distributed (“complete“) or occur in stripes (“partial“). Norrbom (1994) divided the character into microtrichia absent, microtrichia evenly distributed, two states with microtrichia 54 forming stripes, and one state with stripes and bare areas. Although scutal microtrichia is a distinctive feature of a number of species, it is not a suitable cladistic character for species studied here for at least two reasons. First, presence of a medial stripe only in E. canadensis and the position of sublateral and lateral stripes relative to the dorsocentral seta suggest that different patterning systems operate to form scutal stripes (see discussion of tergal pattern systems in section on abdominal characters below). Thus, stripes of E. canadensis may not be homologous with stripes of the remaining species. Second, the absence of microtrichiose stripes in species with microtrichia that do not form stripes is not equivalent to the absence of microtrichiose stripes in species without discal microtrichia. This is actually an amalgam of states from two characters: 1) Presence or absence of microtrichia, and 2) the arrangement of microtrichia (stripes or no stripes). Because absence of microtrichiose stripes depends on the presence of microtrichia, the character is valid only for the subset of species that have discal microtrichia. As noted above, species with the Cerpomya-like scutal pattern have black, velvety microtrichia on the dark pattern elements. This type of microtrichia is found elsewhere only in the R. psalida group where it occurs only on the supra-alar area (Character 15). Homology of the velvety microtrichia in the two groups is uncertain. If the microtrichia are considered an integral part of the Cerpomya-like pattern, then it arose independently in the R. psalida group. If the microtrichia evolve independent of scutal pattern then it could be retained on the supra-alar area and lost elsewhere. Vestiture of the mediotergite (Character 16) varies from sparse, simple microtrichia occurring only laterally (e.g., G. wiedemenni, R. basic/e, R. psalida), to moderately dense, simple microtrichia covering the entire sclerite (most species), or dense pollenose microtrichia throughout (Cerpomya spp. and M. pardalina). Berlocher (1981) refers to the “Polinosity [sic] on the postscutellum“ (=mediotergite + Iaterotergites), which he scores as present in E. canadensis and absent in the other 55 species in his analysis. The microtrichia in E. canadensis are simple and decidedly not like the pollenose microtrichia of Cerpomya and Myioperdelis. Further, at least some microtrichia are present on the mediotergite or Iaterotergites of all the species common to his. and my analyses. Chaetotaxy. Position of the dorsocentral seta has traditionally been used as a key character of the Tephritidae and most, if not all, contemporary taxonomists consider its position taxonomically or phylogenetically important. Hardy (1973, 1974) and Foote (1980) used dorsocentral seta usually behind the supra-alar seta to help separate the Trypetinae (except Adramini) from the Tephritinae (in which the dorsocentral is before or near the supra-alar). Foote and Steyskal (1987) and Foote et al. (1993) used relative position of the dorsocentral to help separate trypetine and tephritine genera. Hancock (1986) and White (1988) considered the position of the dorsocentral relative to the supra-alar to be of importance in the higher classification of the family. Berlocher (1981) divided the position of the dorsocentral seta into three states relative to the supra-alar seta: “slightly behind,“ “far behind,“ and “ahead“ of the supra-alar. , Berlocher (1981) did not specify the distinction between “slighfly' and “far“ behind the supra-alar. Norrbom (1994) used the states dorsocentral seta closer to the level of the postalar seta, and dorsocentral seta closer to the level of the supra- alar seta. In species examined here, position of the dorsocentral seta varies from just behind the transverse suture (e.g., C. schineri, R. psalida group) and well ahead of the level of the postsutural supra-alar seta to near the level of the acrostichal seta and well behind the postsutural supra-alar (e.g., Zonosemate spp.). Further, the ratio of the distance of the supra-alar seta from the transverse suture to the distance of the dorsocentral seta from the transverse suture varies continuously across species (Figure 96; the dorsocentral seta is even with the supra-alar seta when the ratio = 1, behind the supra- alar when the ratio is < 1, and anterior to the supra-alar when > 1). Although the 56 character may be useful in taxonomy, it is phylogenetically uninformative for species studied here. Principal thoracic setae are of uniform color except for the scapulars and proepisternal in some species. Color of the outer scapular seta and proepisternal setae (Characters 18 and 19) varies from whitish or yellowish to dark brown or black; both are polymorphic for a number of species. Berlocher (1981; characters 41 and 48) divided color of thoracic setae into “light“ and “dark“ states. Factors affecting the color of thoracic setae are probably the same as those for head setae (see above). Halter. Halteres are wholly yellowish or brownish, or with the stem yellowish and the knob dark brown or black (Character 17). Brownish halteres occur infrequently in species with yellowish ones and it is likely that the difference in color is a preservation artifact. The bicolored state is distinct from either wholly yellowish or brownish halteres, and occurred only in the R. pomonelle species group, O. beameri and O. nigra. Bush (1966) considered halter coloration diagnostic for the pomonelle group. Norrbom et al. (1988) scored the halter knob as “dark“ (versus “light“) for O. nigra and O. beemeri, and stated that this character suggests a close relationship between the two species. Wing (Characters 21—26, Tables 5, 7) Wing Pattern. A detailed discussion of wing pattern, including the terminology used here, is given in Chapter 2. Band r-m is present (Character 21) in about half of the species examined (e.g., Figures 76, 79—81). In Rhegoletis cerasi, some specimens have the band fused with band dm-cu or the apical band. Two of four specimens of Rhegoletotrypeta rohweri and one of three specimens of Rhegoletotrypeta uniformis lack band r-m. Norrbom (1994) scored the band as present in both of these species, but absent in Rhegoletotrypeta ergentinensis and Rhegoletotrypeta parallela. In R. striatella, there is a very faint mark in cell r1, about where band r-m should occur, but it may be due to an area of dense 57 microtrichia. The anterior portion of band dm-cu may be confused with band r-m in Trypeta inaequalis, but based on a comparison of wing patterns of North American Trypeta (Foote et al., 1993, figures 475—481), band r-m is absent in this species. There is a band passing over vein r-m in the Eulia species (Figure 78), but it is not clear that it is only band r-m. In these species, it appears that band dm-cu anterior to vein M and the proximal end of the apical bands are displaced proximally and have coalesced to form a compound band that may include band r-m. Berlocher (1981, character 13) unknowingly compared proximal band sc of E. canadensis with distal band so of his other species. Norrbom (1989) stated that the “accessory costal band“ (=band r-m) is present in “about half the species of Carpomyina, but rare in other Tephfifidaef The subcostal band crosses vein r-m (e.g., Figure 79) in most species examined (Character 22). It is reduced in Trypeta fracture and T. inaequalis to a pigmented area extending from cell so into cell r1 and a pigmented area surrounding vein r-m. A pigmented spot lying on vein CuA1 is also sometimes present in T. inaequalis. The band is unbroken and well developed in Trypeta tortile (see Foote et al., 1993, figure 476). A hyaline spot is enclosed in band sc within cell br (Character 24) in Acidia cognate, and the Euleia (Figure 78) and Streuzie species. The aberrant wing pattern of S. Iongipennis (see Steyskal, 1986, figure 20; Foote et al., 1993, figure 413) was not included in the character analysis. Members of the R. pomonelle group are distinguished by having bands so, r-m, and dm-cu fused anteriorly, and bands h and sc fused posteriorly (Bush 1966) (Character 25, Figure 82). Anterior fusion of bands also occurs in Rhegoletis zernyi. but bands h and so are free posteriorly. Wing pattern is also unique for species of the Rhegoletis cingulete group (Bush, 1966) (Character 26, Figure 84). Secondary division of the anterior apical band places the posterodistal corner of the anterior arm of the apical band well ahead of the 58 apex of vein M in these species (see Chapter 2). The anterior apical band in other species is either entire or absent. Calypter. Hairs of the calyptral fringe (Character 23) are usually whitish in Rhegoletis, but in the outgroup species they are usually dark brown or blackish, at least proximally. Color of the calyptral fringe is polymorphic for five outgroup and five Rhegoletis species. Venetian and Chaetotaxy. The position of vein r-m and the presence of setae on vein R4+5 are used extensively in tephritid systematics. These characters were not included in the cladistic analysis here for the reasons given below. Bush (1966) considered the position of Mn near the middle of vein M to be one of several diagnostic characters for Rhegoletis. Han et al. (1993) reported that r-m is located beyond the apical 0.40 of cell dm-cu in most genera of the Trypetini. Norrbom (1994) narrowly divided the’ character into two states (<0.60 and >0.63) using the ratio of the length of vein M between veins bm-cu and r-m to the length of M between veins bm-cu and dm-cu. Using Norrbom's ratio, the position of r-m in species studied here is usually near the middle of vein M (mean = 0.52; ). However, position of r-m varies continuously (Figure 97) and can not be objectively divided into discrete states. The average range (0.06) for species where more than one specimen was measured overlaps the difference between Norrbom's (1994) states. Bush (1966) regarded the “setulose“ condition of vein R4+5 to be primitive, and stated that the condition demonstrated an affinity between the Holarctic R. alternate group and most Neotropical species of Rhegoletis. Foote (1981) reported that setae on R4+5 are present almost to crossvein dm-cu in most Latin American Rhegoletis. One or more setae occur on the dorsal surface of vein R4+5 beyond Rs in at least some specimens of most species examined here. Within Rhegoletis the number of setae on R4+5 varies continuously from zero to 15 setae per wing (Figure 98). Number of setae for all species. studied varied continuously from zero to 24 setae per wing (Figure 99). 59 Eighteen (18.2%) of the species had specimens with and without setae. Therefore, the distribution of setae could not be objectively parsed into discrete states, as is required of cladistic characters (Pimentel and Riggins, 1987). Further, the range in the number of setae for many of the species (Figure 99), suggests that there is a large component of individual variation. Although the position of vein r-m and presence or absence of setae on vein R4+5 may have taxonomic utility, they are of no value as cladistic characters in the species studied. Legs. (Characters 27—33, Tables 5, 7) Coloration. Preliminary observations suggested that generally yellowish files have wholly yellowish legs while flies that are generally brownish or black have yellowish legs with dark markings (infuscations). Specimens were initially scored as having infuscate legs if one or more legs had brownish or black markings on one or more segments (excluding the tarsus); or as having wholly yellowish. legs if none of the segments had dark markings. Tarsi were scored separately because it was noticed that infuscation of distal tarsomeres varied independently of the color of the rest of the leg. Initial scores showed that leg coloration is polymorphic for a number of species and is not necessarily correlated with general body color. For example, many specimens in the R. cingulete group, which contains species with generally black bodies, have wholly yellowish legs. Conversely, R. zoqui and R. complete are generally yellowish but often have infuscate legs. Leg coloration also can be sexually dimorphic. Males of R. ribicola have wholly yellowish legs, or if dark markings occur they are limited to the coxae, while females have infuscated coxae and femora. Legs of male R. complete, R. ramosae, and R. zoqui often are more infuscate than females (see Bush [1966] and Hernandez- Ortiz [1985] for detailed descriptions of leg coloration). To determine if there is a pattern to leg coloration, each segment was scored for the presence or absence of infuscation (Table 9). These scores showed that there is no clear 60 transformation from wholly yellowish to wholly infuscate legs, despite ordering the scores by leg or by segment. Infuscation of one segment does not ensure that other segments will be infuscate. The most frequently infuscated segments are the coxae and femora, particularly the hind coxa and hind femur (Table 9). In species with one or more segments infuscated, coxae or femora were not infuscate only in the R. ferruginea species group, Rh. uniformis, and the Zonosemate species; in these species, one or more tibiae (but usually the hind tibia) are infuscate. Based on these observations, it was decided that the hind coxa or hind femur would be used as an indicator of coloration. Infuscation of the hind coxa and hind femur occurred with equal frequency, but, because it is sometime difficult to see the entire hind coxa, the hind femur was chosen (Character 27). Despite limiting variation to this single segment, leg coloration was still polymorphic, although less often than when all segments are taken together. Bush (1966) used coloration of the fore coxa to differentiate R. cingulete and Rhegoletis indifferens. According to Bush (1966), the entire fore coxa is yellowish in R. cingulete while in R. indifferens the posterior surface of the fore coxa is infuscate. Two of the eight specimens of R. indifferens examined here lacked infuscation. Berlocher (1981, character 56) scored infuscation of the fore femur as “complete,“ “restricted to a thin line,“ or “absent.“ In specimens examined here, the amount infuscation of the fore femur varied continuously from none to essentially complete. I found that the fore femur in three of the species Berlocher (1981) scored with complete or restricted infuscation (Rhegoletis carnivore, R. pomonelle and R. ribicola) were polymorphic. I also found that three species Berlocher (1981) scored as having no dark markings on the fore femur were polymorphic (R. complete and Rhegoletis mendax) or had the fore femora infuscate (O. latifrans). Foote (1981, table 1) showed that the amount of infuscation of femora and tibiae overlaps broadly in the Rhegoletis nova group. As with other leg segments, there is considerable intraspecific variation in color of the tarsi (Character 28). One or more legs may have the distal tarsomeres infuscate, 61 and there is no strict correspondence between infuscation of distal tarsomeres and general body color. For example, Rhegoletis electromarpha, R. tabelleria, and R. nr. tabelleria have generally black bodies and heavily infuscated legs, but have tarsi that are wholly yellowish. On the other hand, specimens of S. intermedie, P. immaculate, and several Rhegoletis species that have legs and body generally yellowish, have tarsomere 4 or 5 or both dark brown. Judging color may be influenced by pubescence or a specimen's age, and is generally more of a problem with tarsomeres than other leg segments. Bush (1966) reported dark distal tarsomeres for R. complete, but I found the state more widely distributed among the North American species (Table 5). Foote (1981) states that the tarsi of all legs of Latin American species are yellow, but several of the species examined here had some or all specimens with dark distal tarsomeres. Chaetotaxy. About a third of the species examined lacked a distinct posterodorsel row of setae on the mid tibia (Character 29). (A row is “distinct“ if the setae taming it are definitely larger than surrounding setae. Further, the setae are often semierect, and there is often a bare, narrow strip on one or both sides of the row.) In some species (e.g., R. fausta, R. cerasi, Rhegoletis berberidis), the posterodorsel row of setae can be located, but the setae are questionably different from surrounding setae; such species were scored as not having the row. In other species (e.g., A. cognate, Oedicarena spp.), no posterodorsel row could distinguished. Most species have a distinct anterodorsal row on the hind tibia (Character 30). The row is indistinct or lacking in My. Iimata, S. intermedie, and T. inaequalis, and polymorphic in S. Iongipennis and S. perfecta. Enlarged setae are lacking an the mid or hind femora or both in most species examined (Character 31). In A. cognate, Oedicarena persuase, and Oedicarena tetanops, enlarged setae occur on both femora only in males. Norrbom et al. (1988) scored the posteroventral and anteroventrel femoral setae as “weak“ for 0. persuase and O. tetanops, and considered the setae in these two species to be “no stranger than in many 6 2 other Trypetini.“ Both sexes of O. latifrans and O. nigra have enlarged setae on both femora. Males of P. immaculate have enlarged setae on the mid femur and probably also the hind femur. Setae in Pereterellie superba, P. varipennis, and P. ypsilon intergrade from unmodified to enlarged with a tendency for males to have slightly larger setae than females. Enlarged setae are in indefinite rows on the ventral surface of the mid and especially hind femora in A. cognate males. In the Pereterellie species, enlarged setae are on the posteroventral surface of the mid femur, and are largest in about the distal one-third. Enlarged setae are mostly confined to the posteroventral and anteroventrel surfaces of the mid and hind femora of the Oedicarena species. Enlarged setae also occur ventrally on the fore femur of 0. beemeri, O. letifrons, and O. nigra (see Character 32 below). Enlarged setae in the anteroventrel row of the fore femur occur only in males of My. lucida and the R. ferruginea species group (Character 32). Foote (1981) noted that the “longest and heaviest“ setae of the fore femur occur distally in ferruginea and adusta and medially in blenchardi. The enlarged ventral setae on the fore femur of O. beemeri, O. letifrons, and O. nigra are in addition to setae in the anteroventrel row, which are unmodified. Shape. The fifth tarsomere of R. lycopersella, R. tometis, and one of the three specimens of Rhegoletis acuticarnis is relatively small, cylindrical, and about twice as long as its maximum dorsal width (Character 33). The fifth tarsomere in other species examined is larger, flattened, and less than twice as long as its maximum dorsal width. Abdomen (Characters 34 —35, Tables 5, 7) I restrict the terms tergite and sternite to subdivisions of the sclerotized plate that forms a tergum or sternum, respectively (see McAlpine, 1981a). For example, the basal abdominal tergum (syntergum 1+2) is formed by tergum 1 and tergum 2 and a suture usually can be traced where the sclerites have fused. Thus, the consolidated tergum can be divided into an anterior tergite 1 and posterior tergite 2. 63 Coloration. Terga are either uniformly pigmented or patterned. Uniformly pigmented terga vary from yellowish (e.g., R. alternate species group, Rhegoletis meigeni, O. persuase) to brownish or black (e.g., O. letifrons, R. fausta, dark morphs of Eu. heracler). Tergal patterns consist of dark, regular or irregular shaped marks on predominantly light colored terga, and dark or light colored terga with yellowish or whitish bands along the posterior margin (Figure 100). Species with maculate patterns include E. canadensis, the Pereterellie species, 8. intermedie, and the Zonosemate species. The Rhegoletotrypeta species and most Rhegoletis species have terga with marginal bands; however, both maculate and banded patterns occur in Rh. uniformis, R. caucasica, Rhegoletis chionanthi, R. cingulete, R. camp/eta, R. ferruginea, Rhegoletis flavicincta, R. ramosae, and R. zoqui. The evolutionary relationship between maculate and banded patterns is not known. If wholly yellowish terga and wholly black terga represent extremes of coloration, then maculate and banded patterns may represent intermediate stages, and transformation from one extreme to the other would be by expansion or contraction of pattern elements (Figure 100). An example of intermediate forms in such a transformation can be seen in species where both maculate and banded patterns occur (e.g., see Bush, 1966, figures 49—56; 59—64). Maculate patterns can be further divided into those with dark marks lying on the median line and those with dark marks lying laterally to the median line. The first arrangement of dark marks is here termed the medial pattern system and the second arrangement is termed the sublateral pattern system (Figure 100); both systems are symmetrical about the median line. Markings may be discrete spots on each tergum or may form part of a more extensive pattern, such as stripes running the length of the abdomen (e.g., Cr. tau). I Of the species included in this study, only E. canadensis has the medial pattern system. Out of 34 specimens of E. canadensis examined, the tergal pattern could be 64 completely discerned for 18 (unctuous substances and discoloration partially or entirely obscured the pattern in the other 16 specimens). Out of these 18, all have a dark mark on the median line of tergite 1 and tergite 2, and nine had a dark mark on the median line of one or more additional terga. There is no indication that the marks are a result of the fusion of sublateral pattern elements. Specimens of several Rhegoletis species (ferruginea, caucasica, camp/eta, kurentsovi, magniterebre, zoqui, cingulete, chionanthi, osmanthi) have a medial dark mark on tergite 1 or tergite 2 or both, but, in some of these the mark appears to be the result of the fusion of sublateral marks. Further, a dark medial mark was observed on preabdominal terga distal to tergite 2 only in E. canadensis. The sublateral pattern system occurs in the Pereterellie species, 8. intermedie, and perhaps the Zonosemate species. In the Zonosemate species, the pattern is limited to a pair of lateral (not sublateral) spots on tergum 5 of males and tergum 6 of females. There is some evidence that species with lateral maculae can be derived from species with the medial pattern system or vice versa (see Aczél, 1955a, 1955b, figures 97 and 1 02). Interestingly, within the Tephritidae, many tergal patterns can be grouped into one of these two systems, and there is a strong tendency for only one pattern system to occur in a subfamily. Results of a preliminary survey of the literature and specimens in the Michigan State University Entomological Museum and my personal collection are given in Figure 101 and Tables 10—11. If pattern systems are independently distributed with respect to subfamily then, on average, each subfamily will have equal numbers of species for each pattern system. However, my sample of tergal patterns differs significantly from this hypothesis (Dacinae: x2 = 113.03, p << 0.001; Trypetinae: x2 = 23.15, p << 0.001; Tephritinae: 12 = 47.08, p << 0.001). The distinctiveness of these systems is taken as evidence that two different developmental processes are involved in determining tergal patterns. Similar processes may regulate patterns of 65 microtrichiose stripes on the scutum and pigmentation of the scutellum and mediotergite. Tergal patterns were not used as cladistic characters. As discussed above, maculate patterns may be developmentally different in the medial and lateral pattern systems, and are therefore not strictly homologous. Instead of pattern, ground color of terga (Character 34) was used, but not without some difficulty (see also Norrbom et al., 1988, table 1, character 2). In 0. beemeri, R. complete, and R. suavis the distinction between brownish and yellowish terga is not always clear. In specimens of several species with patterned terga (e.g., Rh. uniformis, R. flavicincta, R. cingulete, Rhegoletis osmanthi), the proportion of light and dark pattern elements is about equal, making the choice of ground color equivocal. The most extreme case of polymorphism is in Eu. heraclei where specimens with wholly black and wholly yellowish terga occur (see also Foote, 1959, p. 149; White, 1988). Norrbom (1994) noted that abdominal color is “highly variable“ in the Tephritidae. Vestiture. Tergal microtrichia intergrades continuously from nearly absent, with only a small amount basolaterally on one or more terga (e.g., E. canadensis, Euleia spp., Chetostoma spp., Pereterellie spp., Zonosemate spp.), to densely microtrichiose (e.g., R. flavigenuelis, Cerpomya spp., M. pardalina). Setae are always present, with the largest ones occurring on the posterior margin of terga. Except for females of E. canadensis, setae along the posterior margin of terga grade from relatively short medially to relatively long laterally. Tergal setae of female E. canadensis are unusual (Table 6) because they are relatively long medially and grade into decidedly shorter setae laterally, especially those on syntergum 1+2 to tergum 4. Excluding tergite 1, tergite 2 plus one or more terga have bands of light and dark colored setae, or the setae are uniform in color (Character 35). The large marginal setae were not scored because they are usually dark colored regardless of the color of other setae. When setae occur in bands, the proximal band is dark and the distal band is 66 light. Bands are sometimes limited to the central area of the terga. Setae are usually concolorous in R. flavicincta, Rhegoletis juniperina, Zonosemate electe and Z. vittigera, but specimens of each species had one or more terga with the banded state. Male Genitalia (Characters 36—62, Tables 5, 7). See Chapter 1 for a detailed description of male terminalia and discussion of the terminology adopted here. Pregenital Segments. The broad, right lateral portion of sternum 7 (Figure 2) in some specimens of R. alternate, R. chionanthi, R. indifferens, and R.,tabellaria, and ‘all specimens of R. cerasi had polygonal surface sculpturing (Character 36); other species lack sculpturing. Most species have one or more setae on syntergosternum 7+8 (Character 46). The character is polymorphic for Rhegoletis berberis and R. cerasi where eight of nine and two of nine specimens, respectively, have setae. A small blister— or sac-like structure (Figure 1) occurs in the intersegmental membrane between sterna 6 and 7 in the Rhegoletotrypeta species and a number of Rhegoletis species. Norrbom (1994) described this structure as a “mostly membranous lobe“ on sternum 6 and found its presence to be a synapomorphy for the Rhegoletotrypeta annulete group. The character was not included in the cladistic analysis here because it was ambiguous for a number of the Rhegoletis species. Epandrium. In the Oedicarena and Pereterellie species, the epandrium is produced posteriorly and the angle formed by its posterior edge below the proctiger and the long axis of the surstyli is less than 90° (Character 39, Figures 9, 12). The subepandrial sclerite and the lateral sclerotizedarms that attach it to the bacilliform sclerites are also shifted posteriorly. In other species (e.g., Figures 7, 16, 23, 27, 29), the epandrium is not markedly produced posteriorly and the angle formed by the epandrium and surstyli is about or more than 90°; the sclerotized arms that attach the subepandrial sclerite to the bacilliform sclerites are more or less vertical. Norrbom et al. (1988) 67 regarded Oedicarena and Pereterellie to be sister taxa based, in part, on the unusual shape of the epandrium and surstyli. The epandrium has numerous evenly distributed microtrichia in most Nearctic and several Palearctic Rhegoletis, Rh. pastranai and Rh. rohweri (Character 45). Other species lack microtrichia or have at most a few distributed randomly. The character is polymorphic in R. suavis and Rhegoletis zephyria. This character appears to be taxonomically useful for separating R. basiola (which has numerous microtrichia) from its sister species R. alternate, and separating R. carnivore (which lacks dense microtrichia) from other R. pomonelle group species. Surstyli. Microtrichia are present at the base of the surstyli anteriorly (Character 43) in a number of Rhegoletis and outgroup species. The character is polymorphic in 15 of the 37 species for which it was recorded. Because of the delicate nature and location of the microtrichia, they may be easily rubbed away during copulation, and therefore some of the polymorphism may be due to scoring artifacts. - The membrane connecting the bacilliform sclerite to the surstylus (Figure 37) has microtrichia (Character 44) in all Carpomyina except Cr. tau. The membrane lacks microtrichia in all other species except Ch. curvinerve. The tips of the surstyli have a cluster of noticeably longer setae in species of the R. cingulete and R. suavis species groups (Character 48). The setae tend to be curved in the R. cingulete group (Figures 19—20) and more or less straight in the R. suavis group. The Chetostoma and Euleia species and R. berberis have one or a few setae at the tip of the surstyli but not a cluster as in the suavis and cingulete groups. Bush (1966) considered the apical surstylar setae to be diagnostic only for the R. cingulete group. Similarly, Berlocher (1981) scored the character as present for R. cingulete and R. indifferens, but not for the suavis group species. Cerpomya species have relatively short, stout, proximally directed, setae on the surstyli distally (Character 49, Figure 13). Setae in this region of the surstyli are not 68 different from adjacent setae in all other species examined. Bush (1966) suggested that Rhegoletis is synonymous with Cerpomya, but could find no “suitable" characters to distinguish the two genera. The surstylar setae described here are unique to the three species of Cerpomya studied. However, before this character is considered a synapomorphy for the genus (see Cladistic Analysis below), the taxonomic status of C. caucasica, the fourth and only other Cerpomya species listed in the current Paleartic Catalog (Foote, 1984), must be established. (V. Korneyev, [pers. comm.] has suggested that C. caucasica is a namen dubium.) Norrbom (1989) considered a surstylus with the posterior lobe elongated beyond the anterior lobe to be a possible synapomorphy for the Carpomyina. Norrbom et al. (1988), Norrbom (1989), and Foote et al. (1993) felt that a similar surstylar shape in Pereterellie indicates a relationship between it and the Carpomyina. In the species studied here, the posterior surstylar lobe varied from absent (Figures7—8, 29—30) to relatively very long (Figures 21—22). Visual estimates indicate that across all species length of the posterior lobe beyond the anterior lobe varies continuously. Further, length of the posterior lobe is similar in species where the epandrium and surstyli are decidedly different in shape. For example, the relative length of the posterior lobe in the Pereterellie and Euleia species is similar to that of several Rhegoletis species, but other aspects of shape vary widely (c.f. Figures 12 and 14). As Norrbom (1989) pointed out for Pereterellie and carpomyines, similarity in shape of the surstylus may be due to homoplasy. Most species have anterior and posterior surstylar lobes, but an additional medial lobe occurs in A. cognate and T. inaequalis (Character 51, Figures 28, 31). Only the anterior lobe is present in E. canadensis and the Zonosemate species (Figures 7—8, 29— 30). Shape of the surstyli is very different between E. canadensis and the Zonosemate species and it seems unlikely that their anterior lobes are homologous. Norrbom (1994, character 37) stated that the anterior lobe (=“mesal lobe“) is absent in 69 Zonosemate, but noted that the rugose apex of the surstylus may be homologous to the anterior lobe; Norrbom (1994) considered absence of an anterior lobe to be apomorphic for Zonosemate. Bacilliform Sclerites. In A. cognata and T. inaequalis, the inner prensiseta is on a relatively large tubercle that places it decidedly distal to the outer prensiseta (Character 54, Figure 35—36). Both prensisetae are at about the same level in most species. In 0. persuasa, O. tetanops and O. nigra, the inner prensiseta is somewhat more distal than the outer prensiseta, but it is not on a tubercle. ln 0. Iatifrons and the Paraterellia species, the prensisetae are at about the same level, but the larger inner prensisetae gives the illusion that it is distal to the outer prensisetae (Figures 9—12). Inner and- outer prensisetae are similar in size in most species (Character 57). The Hhago/etotrypeta species (except Rh. pastranai), R. kurentsovi, and R. meigeni have the inner prensiseta smaller than the outer. The inner prensiseta is larger than the outer in T. inaequalis (Figure 28), and the Myoleja, Oedicarena, and Paraterellia species. Han et al. (1993, 1994a) considered the 'reduced subapical [=outer] prensisetae" to be the synapomorphy for their Trypeta group. The Trypeta group contains eleven mostly Palearctic genera and does not include the genera Myoleje, Oedicarena, and Paraterellia, which also have the reduced outer prensiseta. The shape of the inner prensiseta in Oedicarena and Paraterellia is squarish instead of the usual conical shape, and both prensisetae are mostly or entirely covered by the posterior surface of the bacilliform sclerite. A dorsal keel or ridge is present at least distally on the bacilliform sclerite of 29 of the Rhegoletis species (Character 42). It is usually visible in lateral view and may be erose or serrate (Figure 19). This character was identified only for Fl. berberis by Bush (1966). Presence of the dorsal surstylar keel in R. cornivora appears to be - useful for separating it from other H. pomonella group species, which lack a keel. Anterolateral lobes on the bacilliform sclerites (e.g., Figure 16) are present in all 70 species except those or Oedicarena (e.g., Figure 9), Paraterellia (e.g., Figure 12), and Strauzia (Character 58). One specimen of O. latifrons had a small extension of the anterolateral corners of sternum 10, but it was not a definite lobe (see also Norrbom et al., 1988). Phallus. A basiphallic vesica occurs in Ft. electromarpha, H. ribicola, R. tabellaria, O. latifrans, O. nigra, and the Euleia and Paraterellia species (Character 37). In P. immaculate (Figures 42—43) and P. ypsilon, the vesica is covered with a scale-like pattern. Bush (1966) noted a vesica for R. tabellaria, but does not mention one for R. ribicola. A basiphallic vesica is widespread in the Tephritidae, occurring in Dacinae (Hardy, 1973, 1974; Munro 1984), Acanthonevrini (Condon and Norrbom, 1994), and Tephritinae (Freidberg and Mathis, 1986).- The subapical distiphallic lobe is either trumpet-shaped or forms an elongate lobe (Character 41). Nearctic Chetostoma species have an elongate lobe with large apical hooks (Figure 68) while the other non-Carpomyina species have the trumpet-shaped lobe (Figures 42—49). Species of Carpomyina have an elongate lobe that lacksthe large apical hooks (Figures 50—67). Thelobe is free, at least distally, except in Zonosemata where it is contiguous along its entire length with the parameral sheath (Character 61). The lobe may be bare (e.g., Figure 51), denticulate (e.g., Figure 42), fimbriate with (e.g., Figure 52) or without (e.g., Figure 55) a supernumerary lobe, or microtrichiose (Character 62). Microtrichia range from short and sparse (e.g., Figures 62—63) to long and dense (e.g., Figures 58—59, 70); size and abundance of microtrichia intergrades continuously among species. f Berlocher (1981) reported that an 'apical appendage on [the] aedeagus“ is absent in several species. However, his interpretation of the phallus is incorrect if by 'apical appendage“ he meant subapical lobe: a subapical distiphallic lobe is present in all of the species for which he scored its absence. « The parameral sheath of the distiphallus has polygonal sculpturing in a number 71 species (Character 47). Berlocher (1981) recognized this state as a “scale-like pattern" on the phallus of E. canadensis, and Norrbom (1993) referred to ‘platelike or scalelike sculpture [sic]" for this species. Han (1992) used presence of a pattern of ”narrowly fusiform or oblong cells“ (=polygonal sculpturing) on the “dorsal sclerite' of the distiphallus as the defining character for the Trypetini and also to help delimit the Trypetina. (See Chapter 1 for clarification of Han's “dorsal sclerite“ and “median granulate sclerite.") Foote et al. (1993) reported that the presence of “minute scalelike sculpturing in two areas' of the distiphallus is characteristic of the Trypetina. Variation in markings in this area indicates that there is no qualitative difference between the sculpturing Han (1992) used to define the Trypetini (and Trypetina) and the sculpturing I observed in both Trypetini and non-Trypetini species. Further, the amount of the sculpturing is 'highly variable“ (Han, 1992) and ranges from 'greatly reduced or absent' to extensive (Han et al., 1994b). Eleven species of Trypetinae that have polygonal sculpturing on this area of the parameral sheath, but were not included in Han's (1992) Trypetini, are: E. canadensis. (Figures 48—49), 0. nigra, O. persuasa, O. tetanops, the Paraterellia species (e.g., Figures 42—43), and the Ft. ferruginea species group (e.g., Figure 50). Further, the sculpturing may be widespread within the family (e.g., see Munro 1984, p. 10, and figures 11, 70, 76, and 91). Han (1992) and Han et al. (1993) used presence of a “median granulate sclerite' of the distiphallus to help define the Trypetina and Trypeta group, respectively. Freidberg (1994) assigned Notommoides to the Trypetina based in part on presence of “the median granulation of the distiphallus.“ This denticulate area of the distiphallus occurs in E. canadensis (Figures 48—49), and the Oedicarena (e.g., Figures 44—45) and Paraterellia species (e.g., Figures 42—43), none of which are included in Han's (1992) Trypetina or Han et al's. (1993) Trypeta group. A similar denticulate area is present in Blepharoneura (Condon and Norrbom, 1994, figure 3). A definite acrophallus (Character 53) is present in all Carpomyina except for the 72 Haywardina and Zonosemata species. The acrophallus occurs as two or three sclerotized troughs (Figures 51, 60, 61, 62) or a corrugated plate. -In other species, the gonopores are free distally (e.g., Figures 44—45, 48—49) or there is a single large opening (e.g., Figure 68). Norrbom (1994, figure 4, character 41) described a “sinuous internal tube“ for Cr. parkeri and Cr. tau, which I interpret as the acrophallus (Figure 41). The aedeagus is usually enclosed by the parameral sheath (Character 59). Enclosure of the aedeagus depends on the extent of the appressed flap of the sheath (see Chapter 1). In E. canadensis (Figure 48—49), 0. latifrons (Figure 44—45), and P. superba, about one-half or more of the ventral surface of the distiphallus is not covered by the flap distally, leaving the terminal portion of the aedeagus exposed. ,In the other species of Oedicarena and Paraterellia (e.g., (Figure 44), the distal portion of the distiphallus is not completely closed by the flap, but the aedeagus terminates well within the parameral sheath. In most species, the flap extends the entire length of the distiphallus and encloses the aedeagus (e.g., Figures 46—47, 50—69). The basiphallus has a pair of membranous ventral keels in Cr. tau (Figure 41), H. cuculi, O. nigra, 0. persuasa, O. tetanops and the Paraterellia species (Character 60). Keels are usually located on the distal half of the basiphallus, but in Cr. tau they occur at about midlength. The keel on the left side in the Oedicarena species is expanded into a flat lobe for a distance; keels are delicate and similar from side to side in the other species. Ejaculatory Apodeme. Most species have the distal edge of the ejaculatory apodeme coplanar with the blade, but the edge is flared in C. vesuviana, Eu. fratria, Rhegoletis adusta, Fl. ferruginea, and Fl. ribicola, and some specimens of C. incompleta, C. schineri, Fl. alternate, and R. striatella (Character 38). The flared edge of the ejaculatory apodeme in Fl. ribicola is diagnostic among Nearctic Rhegoletis species (Bush, 1966). Hypendrium. The intrahypandrial membrane anterior to the lateral arms of the aedeagal guide varies from unmodified, being more or less tightly stretched across the hypandrium, to forming a sac of varying depth. The hypandrial sac in the R. suavis 73 group is lined with numerous heavily sclerotized denticles (Character 40); when present in other species, it lacks denticles. Bush (1966) found the sac to be diagnostic for the suavis group. Berlocher (1981) listed an hypandrial sac only for suavis group species but it is also present in Fl. striatella (Bush 1966), which was included in his analysis. A hypandrial sac may be widely distributed within the Tephritidae (see Chapter 1). The hypandrial apodeme is absent in the Carpomyina, My. lucida, T. inaequalis, and the Strauzia species (Character 50). In some specimens lacking the apodeme, the hypandrium is thickened medially, but such thickening is not a hypandrial apodeme. Norrbom et al. (1988) considered the apodeme to bepleisiomorphic in Oedicarena and Paraterellia, and noted that it is “weak or absent“ in Rhegoletis and a number of other trypetines. Pregonites. The right pregonite is displaced farther ventrally than the one on the left (Figures 3, 5) in all species except the Chetostoma species (Character. 52). In Chetostoma, the pregonites are more or less symmetrical from side. to side. Proctiger. The hypoproct forms a single sclerotized plate on the ventral surface of the proctiger, except in species of Oedicarena and Paraterellia. In Oedicarena (e.g., Figures 10—11) and Paraterellia, the hypoproct is divided medially (Character 55). In A. cognata (Figure 27), the Chetostoma, Euleia, Myoleja, and Strauzia species, and T. inaequalis (Figure 32), the hypoproct extends dorsally for most or all of the height of the proctiger (Character 56). In other species, the hypoproct extends dorsally for less than half of the height of the proctiger. Female Genitalia (Characters 63—77, Tables 5, 7) Spermathecae and Spermathecal Ducts. Number of spermathecae (Character 63) and number of sperrnathecal ducts (Character 65) each vary from two to four. Species with two spermathecae have two or three sperrnathecal ducts; species with three spermathecae have three ducts; and species with four spermathecae have three or four 74 ducts. The number of spermathecae is usually constant for a given species. However, in some species that usually have two spermathecae and three spermathecal ducts, the distal end of the duct that is typically undifferentiated is sometimes dilated or there is a partially sclerotized, presumably nascent, spermatheca (e.g., R. basic/a, H. chionanthi, Rhegoletis batava, R. flavigenualis, Fl. fausta, R. meigeni). Nascent spermathecae were not included in the number of spermathecae for a given species. In the Oedicarena species there are four spermathecae, and in 0. persuase and 0. tetanops there are four spermathecal ducts. In 0. latifrons and 0. nigra, however, there are only three ducts, one of which is forked distally and bears two of the spermathecae. - Two specimens of Fl. mendax and one of E. canadensis had two spermathecae attached to the end of a forked duct, similar to O. latifrans and O. nigra. Number of spermathecae is often used to diagnose higher taxa. For example, dacines typically have two spermathecae (Hardy, 1973, 1974; Munro, 1984), trypetines have three, and tephritines two (Hancock, 1986; White, 1988). Contrary to the expectation for trypetines, fewer than half (39/88 = 44.3%) of the species examined here have three spermathecae, while slightly more than half (45/88 = 51.1%) have two. Bush - (1966) used number of spermathecae and spermathecal ducts to help diagnose Nearctic species groups of Hhagoletis. Norrbom et al. (1988) considered the four spermathecae of Oedicarena to be a synapomorphy for the genus and also noted that having four spermathecae is unique within the Tephritidae. Norrbom (1989) stated that presence of three spermathecae is pleisiomorphic for the Tephritidae, and used number of spermathecae in his analysis of Latin American carpomyines (Norrbom 1994). Berlocher et al. (1993) considered three spermathecae synapomorphic in the R. pomonella group, but also noted that number of spermathecae in Rhegoletis is likely homoplasious. Number of spermathecae and spermathecal ducts can be diagnostic at the species » level. In the morphologically homogenous R. pomonella group, R. cornivora has two 75 spermathecae (contrary to Bush [1966] and Berlocher [1981]) and three ducts while the other species have three spermathecae and ducts. The sister species Fl. alternate and R. besiole have two and three ducts, respectively. Shape of spermathecae (Character 64) varies from small and spherical (e.g., Myoleje limeta) to long and cylindrical (e.g., Fl. suavis group). Cylindrical spermathecae are straight (e.g., Fl. pomonella) or convoluted (e.g., R. ribicola). In several of the species with three spermathecae, some or all specimens have one spermatheca decidedly smaller than the other two (Character 74); Bush (1966) noted- this size difference for the Fl. pomonella group and R. zoqui. The external surface of spermathcae is smooth or has wrinkles, bumps or variously shaped papillae. Very small, delicate, and essentially colorless, capitate bodies (Ming, 1989, figures 43—44) also are present. Spermathecae may or may not have a definite neck to which the spermathecal duct attaches. An atrium is present where the duct joins the spermatheca in several of the species. Atria may be distinct, sclerotized structures, but identifying them can be quite arbitrary because the place where the duct attaches to the spermatheca is sometimes dilated, sometimes sclerotized, and sometimes sclerotized and dilated. Spermathecal shape across all species appears to vary continuously. Shape within species usually varies to a lesser extent, but in R. tabellaria, spherical, pyriform, and cylindrical spermathecae were encountered. Mating and oviposition may affect spermathecal shape (White, 1988). Because of the difficulty of precisely determining shape by visual estimate, shape was arbitrarily classified into two broad categories: globular and cylindrical. A spermatheca is globular when its body is less than 3 times as long as its greatest diameter, and cylindrical when its body is more than 3 times as long as its greatest diameter. Unfortunately, this dichotomy does not capture the diversity of shapes and is probably not very phylogenetically informative. For example, spermathecae of the Zonosemate species are weakly sclerotized (considered synapomorphic by Norrbom [1990, 1994]). and in Rhegoletis converse, Fi. nova, and 76 the R. psalida group spermathecae have a sharp bend or coil proximally. Berlocher (1981) divided spermathecal shape into “globular,“ “oval,“ and “long and cylindrical,“ but he did not specify the difference between globular and oval. Spermathecal ducts are usually about as long as the abdomen, but in several species they are distinctly longer (e.g., Fl. berberidis, P. kurentsovi, Rhegoletis mango/ice, Fl. psalida, S. intermedie, S. perfecta), and across all species may vary continuously. Ducts are usually membranous and distinct from the spermatheca; however, in a few species (e.g., R. blanchardi, P. meigeni, H. juniperine) some specimens have ducts that are sclerotized for a variable length beyond their attachment to the spermatheca. In R. psalida, the ducts are sclerotized for a short distance beyond the ventral receptacle. Ducts are usually colorless, but in the some Neotropical Rhegoletis (e.g., Fl. nova group) they are lightly pigmented. The external surface of ducts (Character 75) is almost always smooth, but it is definitely annulated in the Zonosemata species. Eversible Ovipositor Sheath. In most species the eversible ovipositor sheath is subequal in length to segment 8 (Character 66). In E. canadensis, O. persuasa, and 0. tetanops, however, the sheath is decidedly longer than segment 8. Norrbom and Kim (1988) reported that “scales“ (=denticles) are absent from the apical portion of the eversible ovipositor sheath of Rhegoletis. Evidently, they were referring to the large discal denticles, because minute denticles are present at the extreme apex of the sheath in Rhegoletis, as well as the other species examined. Denticles on the sheath just proximal to the point of its attachment to segment 8 have either a single point or multiple points (Character 68). These denticles usually are similar dorsally and ventrally, but in the Euleia species and R. cingulete, ventral denticles have single points and dorsal denticles have multiple points. In Fl. ribicola and R. batava, there is a mixture of single and multiple point denticles both dorsally and ventrally. This character is taxonomically useful in the R. pomonelle species group where Fl. carnivore has denticles with single points and the Other species have denticles ‘ 77 with multiple points (Figures 102—103). The large discal denticles on the ventral surface of the sheath usually are triangular and have a single point, but in the Chetostoma and Myoleje species they are squarish and irregular apically (Character 69). Han (1992) regarded the denticles of these genera to be a reduced form of the normal triangular-shaped teeth and a synapomorphy for his Chetostomina. The eversible ovipositor sheath usually bears only denticles, but several species have microtrichia proximally (Character 67). The dorsal taeniae of the eversible ovipositor sheath (Steyskal, 1984; Norrbom and Kim, 1988) usually end well ahead of segment 8. In the Chetostoma and Rhegoletotrypeta species (except pastranai) and My. lucida, however, they reach segment 8 (Character 76). Longer taeniae are correlated with a laterally compressed segment 8 (see below) in the Chetostoma species, My. lacida, and Rh. annulate. Segment 8. The tip of segment 8 is dorsoventrally flattened (wider than high) in most species, but in several of the outgroup species it is laterally compressed (narrower than high) (Character 71). Segment 8 is constricted at its base in several of the outgroup species (Character 70), and there is a tendency for this condition to occur in species where the tip is laterally compressed. The cloaca may be glabrous or surrounded by microtrichia, denticles, or both (Character 72; see Stoffolano and Yin [1987], figure 32). The tip of segment 8 may bear serrations (e.g., Streuzie spp.), subapical points or lobes (e.g., E. canadensis, R. caucasica, Zonosemata spp.), or both (e.g., G. wiedemenni, 0. nigra, R. nova). The tip is usually armed in the outgroup species, but usually bears only a single apical point in most Rhegoletis species (Character 73). Han (1992) used “lateral serrations toward apex“ of a dorsoventrally flattened segment 8 as a . synapomorphy for his Trypetina. However, these states are of questionable phylogenetic importance as both states occur in species not in Han's Trypetina (e.g.,. G. wiedemenni, Cr. tau, 0. nigra). Foote et al. (1993) also used the finely serrate tip of segment8 to 7 8 help diagnose their Trypetina. Norrbom (1994) used several attributes of the tip of segment 8 in his analysis of Cryptodacus, Heywardine. and Rhegoletotrypeta. Throughout the Tephritidae, the tip of segment 8 often has points, projections, or serrations (e.g., see Hardy, 1973, 1974; Stoltzfus, 1977; White, 1988; Condon and Norrbom, 1994; Merz, 1994). Shape of the tip can vary considerably among species within higher taxa, and even within seasonal morphs of a single species (Jenkins and Turner, 1989). Selection on the shape of the tip of segment 8 may result in convergence because of its importance for placing eggs into host tissue. Evidence for convergence may be that similar shapes occur in widely divergent species. For example, Shape of the tip in the dacine Decus deceptus (Hardy, 1974, figure 28a) is essentially identical to that of the tephritine Tephritis baccheris (Jenkins and Turner, 1989, figure 13c). If tip shape is under selection, then occurrence of similar shapes in more closely related species could also be due to convergence. The two states used here (armed and unarmed) are arbitrary and probably not very informative because of their breadth. However, accounting for all variation in shape would result in a large number of states of questionable homology. The only instance of polymorphism was in the specimens of H. ferruginea examined: two specimens had a single apical point and one had a pair of minute subapical points. Berlocher (1981) divided ovipositor shape into “trident shaped“ tip for E. canadensis and “spear shaped“ tip for the other species analyzed. However, Berlocher (1981) did not differentiate between the ovipositors of O. Ietifrons and Z. electe, which have subapical points (see Bush, 1965, figure 18; Norrbom et al., 1988, figure 6a), and species with unarmed tips. Syntergosternum 7. Norrbom (1989, figure 6) defined the Carpomyina by the presence of a weakly sclerotized area at the apex of syntergosternum 7. The area is usually apparent in carpomyines with a darkly pigmented syntergosternum 7, but it is sometimes‘difficult to locate in lightly pigmented specimens. In lightly pigmented flies, 79 it is sometimes possible to recognize the weakly sclerotized area by using phase contrast microscopy (300x) and comparing the cuticle from areas immediately encircling the insertion of setae with areas lying between setae. The area encircling the base of a seta often forms a small sclerotized “island“ that serves as a reference for comparing areas lying between setae. The weekly sclerotized area is present if the areas lying between setae are largely unsclerotized. A weekly sclerotized area almost always occurred in species of Carpomyina examined; however, its presence was ambiguous in a number of specimens even after re-examination. Further, I was unable to identify any weakly sclerotized area in the Zonosemate species. Therefore the character was not included in the cladistic analysis. Setae of syntergosternum 7 are usually unmodified, but in several species the tip has about 8-16 stout setae ventrally (Character 77). Cladistic Analysis “...in all fields of biology one needs, now and then, to ask whether current theory, however satisfying, provides a clear view of reality.“ — Evans (1977) With polymorphisms excluded from the data set, 18,691 most parsimonious reconstructions (MPRs) were generated before insufficient computer memory ended the search. These trees were 135 steps in length and, excluding uninformative characters, had a consistency index (CI) of 0.417 and retention index (RI) of 0.748. The trees are summarized by the consensus cladogram in Figure 104. When all taxa were included in the data set (Figure 105), the Cl increased to 0.430 and the Rl to 0.848. Reweighting the data set (Ferris, 1969; Carpenter, 1988) using the rescaled consistency index (RC) of the 18,691 trees (RC = 0.322) resulted in 12,061 MPRs with a CI of 0.671 and an RI of 0.900 (Figure 106). Results from the reweighted data should be viewed with caution because iterations 2—5 of the reweighting routine could not be completed due to excessive computing time or insufficient computer memory or both. None of the 18,691 trees initially generated or the 12,061 trees from the reweighting procedure 80 were compatible with a useroefined constraint tree where Rhegoletis was placed in a monophyletic clade. A random addition search using the constraint tree produced 1,500 trees 141 steps long; with uninformative characters removed, the CI and RI of these trees were 0.399 and 0.729, respectively. When polymorphisms were included in the data set, 13,100 MPRs (summarized in Figure 107) were found before the search was aborted because of insufficient computer . memory. Length of the trees generated was 306 steps and the CI and RI were 0.281 and 0.789, respectively. None of the 13,100 trees were compatible with the constraint tree filter. Searches with and without polymorphic characters each appear to have found a single tree island as indicated by Rls greater than 0.67 (Maddison, 1991). Although not all trees in either search were recovered, the existence of one tree island for each suggests that shorter trees will not be found with this data set. The monophyly of Rhegoletis was not confirmed whether polymorphisms were included or excluded from the search, or when the data were reweighted. With polymorphisms excluded, six additional steps (135 versus 141) were needed to prodtice trees compatible with the constraint tree. The monophyly of Rhegoletis has been in question (Norrbom, 1989; Foote et al., 1991) and recent analysis of molecular data (McPheron and Han, submitted; Smith and Bush, in review) also indicates that the genus in not a natural grouping. Intergeneric relationships have tended to be poorly resolved for tephritid flies in recent phylogenetic analyses (Norrbom, 1994; Han and McPheron, 1994; Han and McPheron, submitted; Smith and Bush, in review). In this study, carpomyines and non- carpomyines were placed in separate clades in all searches. Intergeneric relationships within each clade were largely unresolved when polymorphisms were excluded (Figure 104) and only slightly more resolved when polymorphisms were included (Figure 107) or the data were reweighted (Figure 106). In all cases, monophyly of the carpomyines 81 was supported by a single character: presence of an elongate subapical lobe on the distiphallus (Character 41). Interestingly, the subtribe's putative synapomorphy (Norrbom, 1989) was not included in the analysis (see discussion of female syntergosternum 7 in Character Analysis). Reweighting the data produced clades not found in other analyses, but, as mentioned above, results from that search should be viewed with caution. In the reweigthed search, Character 65, number of spermathecal ducts, supported the monophyly of a clade containing R. alternate, R. kurentsovi, and the Neotropical species of Rhegoletis (Figure 106). This is of interest because Bush (1966) regarded the R. alternate species group and most Neotropical Rhegoletis to be closely related. Bush based this relationship on a “setulose“ vein R4+5, wing pattern, and head shape, all characters he regarded as primitive. However, as shown in Figure 98, the number of setae on R4+5 varies continuously; head shape and wing pattern were unspecified by Bush and therefore could not be evaluated. Even so, primitive characters are phylogenetically uninformative (Wiley, 1981). Oedicarena and Paraterellia, genera regarded to be closely related to Rhegoletis or other carpomyines (Berlocher, 1981; Berlocher and Bush, 1982; Norrbom et al., 1988; Norrbom, 1989, 1994; Foote et al., 1993), were consistently placed as sister taxa in a clade within the basal polytomy (Figures 104—107). Monophyly of the clade was supported by shape of the epandrium (Character 39; also noted by Norrbom et al., ‘ 1988) and the divided hypoproct (Character 55). Foote et al. (1993) did not place these genera in a tribe, but results of this analysis indicate that they may belong in a group that includes the Trypetina plus Chetostoma and Myoleje species. As pointed out in the discussion of male and female genital characters (see Character Analysis), Han's (1992) Trypetini and Trypetina do not include all species possessing his synapomorphies for the tribe and subtribe. Evidently, Han (1992) was able to narrowly define the Trypetini and Trypetina by not examining many of the species 82 previously placed in the tribe. Based on the observations and analyses reported herein, Han‘s Trypetini and Trypetina and Foote et al.'s (1993) Trypetina are paraphyletic. Myoleje and Chetostoma species also were consistently placed together in a clade in the basal polytomy. Monophyly of the clade was supported by the shape of the discal teeth on the eversible ovipositor sheath (Character 69, Figures 104—107). Han (1992) considered this character to be an unequivocal synapomorphy for his subtribe Chetostomina, which includes the Chetostoma and Myoleje species studied here. A clade containing A. cognate, T. inaequalis, and the Chetostoma, Euleia, Myoleje, and Streuzie species was supported by Character 56 (extent of the hypoproct) with the data reweighted (Figure 106). This grouping corresponds to Han's (1992) Trypetina + Chetostomina and Foote et al.'s (1993) Trypetina + unplaced Trypetini. Within this larger clade, a hyaline spot in wing cell br (Character 24, Figure 106) was synapomorphic for A. cognate and the Euleia and Streuzie species. All three genera are placed in the Trypetina by Han (1992) and Foote et al. (1993); A monophyletic group containing M. pardalina and the Cerpomya species is supported by the dense pollenose microtrichia on the mediotergite (Character 49, Figures 104—107). The unusual short, stout, proximally directed distal setae on the surstyli of. the Cerpomya species are likely a synapomorphy for the genus (see Character Analysis). A clade containing the R. cingulete and R. suavis species groups was supported by Character 48 with the data reweighted (Figure 106). Recent molecular studies (Ming, 1996; Smith and Bush, in review) also support this relationship (but, see Berlocher and Bush [1982] and McPheron and Han [submitted]). Because Rhegoletis is very likely paraphyletic, a single outgroup will not exist for all species currently placed in the genus. The outgroup of the clade containing the Rhegoletis species is likewise uncertain. When polymorphisms were included in the search, a potential outgroup clade containing Cr. tau, the Heywardine and Zonosemata 83 species, and three of the four Rhegoletotrypeta species was found for the remaining carpomyines (Figure 107). However, this clade is weakly supported by the data. There is no synapomorphy SUpporting the clade, and the clade collapses in trees one step longer. Further, the fit of the characters to trees generated with polymorphic characters is decidedly worse than to trees generated with polymorphisms excluded (Cls = 0.281 and 0.430, respectively). However, the decision to accept this clade as an outgroup to the remaining carpomyines centers on the larger issue of using polymorphic characters to infer phylogenetic relationships (Nixon and Wheeler, 1990; Davis and Nixon, 1992; see also Doyle, 1992). The essential problem of using polymorphisms for reconstructing phylogeny is one of ancestor-descendant relationships (Nixon and Wheeler, 1990; Davis and Nixon, 1992). In sexual organisms, polymorphisms that result from recombination are not hierarchic among individuals and, therefore, do not reflect historical relationships (Nixon and Wheeler, 1990; Davis and Nixon, 1992). This does not include age- and sex-specific polymorphisms that are not altered by recombination, which can be phylogenetically informative (Nixon and Wheeler, 1990; Davis and Nixon, 1992). Further, assuming that multiple states of an attribute in a terminal taxon are a result of cladogenesis is incorrect if any of the states are a result of anagenesis (Platnick et al., 1991» Accepting polymorphisms as legitimate cladistic characters means that they can also serve as synapomorphies, and this may present serious problems. For example, Han (1992) defines his subtribe Trypetina by the “...following combination of characters: 1) dorsal sclerite of distiphallus usually with pattern of narrowly fusiform or oblong cells...; 2) distiphallus usually with median granulate sclerite...; and 3) aculeus wide and dorsoventrally flattened, usually with lateral serrations toward apex...“ (reference to Han's figures omitted; emphasis added). By “usually“ Han presumably means that a character may sometimes be lacking, and therefore polymorphic. Thus, inclusion in 84 Han's Trypetina is possible when none, one, two, or all three characters are absent. The grouping becomes quite meaningless. If polymorphisms are to be used at all, methods that are consistent with cladistic theory need to be developed for when they are used as synapomorphies. Qualifiers like “usually,“ “often,“ “generally,“ and “rarely,“ often used in character descriptions, can not exculpate polymorphisms because states of cladistic characters must be mutually exclusive (Pimentel and Riggins, 1987). Polymorphisms also present problems for coding data (Pimentel and Riggins, 1987; Nixon and Davis, 1991; Platnick et al., 1991; Maddison, 1993), and computerized parsimony analysis (Platnick et al., 1991; Maddison and Maddison, 1992; Swofford, 1993). The two most common ways polymorphic characters are coded are as missing data or by assigning them the ancestral state (Nixon and Davis, 1991). However, both methods may lead to erroneous results (Pimentel and Riggins, 1987; Platnick et al., 1991). Maddison and Maddison (1992, p. 48) suggested including a third state called “polymorphic“ for species possessing both states of a binary character. However, this tactic will reflect descent only if the character is truly polymorphic (i.e., every individual possesses both character states). Computer programs such as PAUP currently cannot treat polymorphic characters in a population-genetics sense (Swofford, 1993). These algorithms instead treat polymorphic terminal taxa as groups of monomorphic subtaxa and assigns their ancestor the state that minimizes tree length (Swofford, 1993). Is it possible to accept characters that are only “slightly“ polymorphic while rejecting others that are “highly“ polymorphic? The arbitrariness of such divisions is apparent, but what if polymorphisms are distributed among species in such a way that a gap clearly separates characters that are polymorphic in a few species from characters that are polymorphic in many? Although this appears to give a rational basis for including polymorphisms, there is no logical reason to suppose that characters that are polymorphic within a few species are more phylogenetically informative (i.e., “more 85 fixed“) than characters that are polymorphic in numerous species. In practice, distinctions between characters that are polymorphic in a single species and those that are polymorphic in many species may be purely arbitrary (Figure 108). It may be argued that eliminating polymorphisms from cladistic analyses reduces the phylogenetically informative data available to systematists. An alternative viewpoint, and the one I believe to be correct, is that removing polymorphisms reduces error. Characters that do not reflect descent cannot contribute to the understanding of phylogenetic relationships. Therefore, for all of the reasons discussed above, results of the search using polymorphisms should be viewed with skepticism. We are thus left without a definite outgroup for the clade containing the Rhegoletis species. Despite an extensive character analysis, few phylogenetically informative characters above the species level were found. This is similar to the findings of other recent studies (Norrbom, 1994; Han and McPheron, 1994; Han and McPheron, submitted; Smith and Bush, in review). How then should we view current ' ' classifications of the family (e.g., Hardy, 1973; Hancock, 1986; Foote et al., 1993)? Do these intuition-based arrangements present a “clear view of reality,“ or should we expect polytomies to be common in a “relatively recent, rapidly radiating group“ (Foote et al., 1993)? These questions will only be answered by additional analysis. The character analysis reported herein, although extensive, deals only with the descriptive anatomy of the flies. Future morphological investigations will need to go beyond this preliminary phase of study to discover new characters. Identifying homologies and stabilizing terminology are areas of tephritid morphology that especially need attention. Well-reasoned character analyses are also needed. If our goal is to understand the evolution of fruit flies, then the depth of our knowledge depends on the methodology we choose. “You don 't know what you know, until you know what you don't know.“ — Anonymous SUMMARY The family Tephritidae includes over 4,200 described species (Foote et al., 1993), and contains some of humankind's most important agricultural pests (White and Elson- Harris, 1992). Fruit flies have been the focus of numerous biological studies, especially in the areas of evolution (see Bush, 1992) and behavior (see Jenkins, 1990). Despite the importance of tephritid flies to human welfare and for scientific study, classification of the family has changed little since Hering's classification was published in 1947. Current classifications are untested, intuition-based arrangements of taxa. This presents a significant problem for biologists interested in working with fruit flies because classifications, insofar as they reflect phylogeny, provide the basis for many comparative studies (Miles and Dunham, 1993). The most widely accepted method of inferring evolutionary relationships is phylogenetic systematics (Forey et al., 1992; Kluge and Wolf, 1993). Phylogenetic systematics is a two step process consisting of separate character and cladistic analyses. Characters are identified during character analysis, the extent of their variation determined, and hypotheses about their homology tested. These characters are then used during the cladistic analysis to infer phylogenetic relationships. In phylogenetic systematics, deductive testing can occur only during the character analysis. Cladistic analysis is an inductive method (Bryant, 1989) and, as a result, any cladogram can be explained post hoc. Therefore, confidence in a phylogeny depends directly on the characters used to infer it. Little, if any, attention is usually given to character analysis in phylogenetic studies of the Tephritidae. One problem is that characters useful in taxonomy are often assumed to be phylogenetically informative. 86 87 Taxonomic characters may vary continuously and yet be useful for delimiting taxa in different areas of the range of variation. However, cladistic characters must be discrete (Pimentel and Riggins, 1987). Taxonomic characters may be autapomorphous or homoplasious and still be useful taxonomically. However, autapomorphies and homoplasy are not phylogenetically informative (Wiley, 1981). Polymorphisms may be taxonomically useful when qualified by terms such as “usually.“ However, such qualifiers can not exculpate polymorphisms for cladistic analysis. Useful taxonomic characters, therefore, are not necessarily useful cladistic characters. The male genitalia and wing patterns of fruit flies provide many characters that are useful in taxonomy. Problems in using these characters for phylogenetic analysis include unstable nomenclature and questionable homologies. To improve this situation, a detailed description of the male genitalia of trypetines is given in Chapter 1. The description uses current terminology and homologies for the Diptera, and should be widely applicable within the family. Internal structure of the distiphallus is as yet uncertain, but the ground plan of the phallus proposed in Chapter 1 provides a basis for homologizing distiphallic structures. Chapter 2 presents a system of structural landmarks for identifying wing pattern elements, thereby stabilizing pattern terminology. This system helps ensure that homologous bands are recognized. A heuristic model of wing pattern evolution also was developed in Chapter 2. The model provides a basis for constructing transformation series that can be used to constrain cladistic searches. An example of a transformation series is given for several species of Rhegoletis. The monophyly of Rhegoletis and intergeneric relationships within the Trypetini are examined in Chapter 3. Despite an extensive character analysis, most relationships among the genera remain unresolved. Results of the study indicate that Rhegoletis is not monophyletic; that the Trypetina is paraphyletic; that carpomyines are more derived than trypetines; and that previously unplaced genera are more closely related to 88 trypetines than carpomyines. These findings are contrary to current classifications (Foote et al., 1993). To what extent, then, will it be possible to recover the evolutionary history of tephritid flies? Will our phylogeny of the family be a dichotomous “ladder“ or a polytomous “bush?“ We would all like it to be a ladder: ladders are more certain and inherently more informative. But will we have failed if we can resolve only a bush? Which presents the clearer view of reality? Somewhere between essentialism and nominalism lies truth; knowing what we do and do not know is the art of science. APPENDIX Table 1. Specimens examined. 89 Number of specimensa Species Males Females n Acidiaoognata 3(1) 2(1) 5(2) Cerpomya incomplete 3(2) 4 (2) 7(4) C. schineri 5(3) 5(3) 10(6) C.vesuviane 1(1) 1(1) 2(2) Chetostoma californicum 3( 1) 3(1) 6(2) Ch. curvinerve 2(1) 2(1) 4(2) Ch. rubidium 1(1) 2(1) 3(2) Cryptodacus tau 2(2) 4(2) 6(4) Epochra canadensis 7 (3) 7 (3) 14(6) Euleia fratria 4(1) 3(1) 7(2) Eu.hereclei 1(1) 3(1) 4(2) Eu. uncinata 2(1) 2(1) 4(2) Goniglossum wiedemenni 2 ( 1) 3 ( 1 ) 5 (2) Heywardine cuculi 2(2) 2(2) 4(4’) H. cuculiformis 1(1) 2(2) 3(3) Myioperdelis pardalina 3 (2) 4 (2) 7 (4) Myoleje Iimata 3 ( 1) 3 ( 1) 6 (2) My.Iucida 1(1) 1(1) 2(2) My. nigricornis 1 (1 ) 0 ( 0) 1 (1 ) Oedicarena beameri 0 ( 0) 1 (0) 1 (0) O.latifrons 5(2) 1(1) 6(3) O.nigra 3(1) 3(1) 6(2) O.persuasa 3(1) 1(1) 4(2) O.tetenops 3(1) 3(3) 6(4) Paraterellia immaculate 4 ( 1 ) 3 ( 1 ) 7 (2) P. superba 1(1) 5(1) 6(2) P. varipennis 1(1) 2(1) 3(2) P. ypsilon 2(1) 2(1) 4(2) Rhegoletis acuticarnis 2(1) 1(1) 3(2) R.adusta 1(1) 1(1) 2(2) R. almatensis 1 (1) 1 (1) 2(2) R. alternate 12(8) 8(4) 20(12) R. a. orientalis 1(0) 1(0) 2(0) R.basiola 12(8) 8(4) 20(12) R.batava 3(1) 5(2) 8(3) R. berberidis 6(6) 9(4) 15(10) R. berberis 13(9) 8(4) 21(13) R. blanchardi 2(2) 4(3) 6(5) R.boycei 10(5) 6(4) 16(9) R. caucasica 1(1) 1(1) 2(2) R. cerasi 13(9) 8(4) 21(13) R. chionanthi 9(5) 9(4) 18(9) R. cingulete 10(6) 9(5) 19(11) R. complete 12(8) 8(4) 20(12) R. converse 5(3) 5(4) 10(7) R. carnivore 12(7) 7(4) 19(11) R. ebbettsi 0(0) 1(0) 1(0) R. electromarpha 8(4) 8(4) 16(8) Table 1 (cont'd). 90 Number of specimensa Species Males Females n R.emlliae 0(0) 1(1) 1(1) R.tauste 10(6) 8(4) 18(10) R. ferruginea 3(1) 3(3) 6(4) R. flavicincte 3 ( 1 ) 3 ( 1 ) 6(2) R. flavigenuelis 3(1) 2(1) 5(2) R. “florida' 8(4) 8(4) 16(8) R. indifferens 12(8) 9(5) 21(13) R. jemaicensis 2(1) 5(2) 7(3) R.juglandis 10(6) 8(4) 18(10) R.juniperina 14(10) 12(8) 26(18) R. kurentsovi 3(1) 3(2) 6(3) R. lycoperselle 6(3) 9(4) 15(7) R. macquarti 1(1) 4(2) 5(3) R. magniterebre 4(2) 2(1) 6(3) R.meigeni 10(6) 8(4) 18(10) R.mendex 10(6) 9(5) 19(11) R.metallica 0(0) 1(1) 1(1) R.mongolice 1(1) 1(1) 2(2) R.nova 9(5) 7(5) 16(10) R.absoleta 1(0) 0(0) 1(0) R.csmanthi 8(4) 8(4) 16(8) R.penela 1(1) 0(0) 1(1) R.persimilis 7(4) 12(7) 19(11) R.pomonella 15(11) 12(9) 27(20) R.psalide 7(4) 6(3) 13(7) R.ramosae 1(1) 1(1) 2(2) R.reducta 0(0) 3(0) 3(0) R.rhytlda 1(1) 1(1) 2(2) R. ribicola 12(8) 8(4) 20(12) R. scutal/ate 1(0) 0(0) 1(0) R. striatella 12(8) 9(5) 21(13) R.suevis 15(11) 8(4) 23(15) R. tabelleria 12(8) 7(5) 19(13) R. nr. tabelleria 5(1) 7(3) 12(4) R. tomatis 5(3) 3(2) 8(5) R.turanica 1(1) 0(0) 1(1) R. zephyria 12(8) 8(6) 20(14) R.zernyi 1(1) 1(1) 2(2) R.zoqui 6(4) 8(4) 14(8) Rhegoletotrypeta ennuleta 1(1) 4(1) 5(2) Rh. pastranai 1 (1) 1 (1) 2(2) Rh. rohweri 2(1) 2(1) 4(2) Rh. uniformis 1 (1) 2(1) 3(2) Streuzie intermedie 3 ( 1) 3 ( 1 ) 6(2) 8. Iongipennis 3 ( 1 ) 3 ( 1 ) 6(2) S.perfecta 3(1) 3(1) 6(2) Trypeta fracture 0(0) 1(0) 1(0) T. ineegla/is 5(2) 7(4) 2(2) Table 1 (cont'd). 91 Number of specimensa Species Males Females n T. tortile 0(0) 1(0) 1(0) Zonosemate electe 3 (2) 3 (2) 6 (4) Z.scutellata 2(1) 1 (1) 3(2) 2. vidrapennis 1 ( 1 ) 1 ( 1 ) 2 (2) Z.vittigera 3(2) 3(2) 6(4) TOTAL 462(278) 417(227) 879(505) aNumbers in parentheses are the number of specimens for which genitalia were examined. 92 Table 2. Comparison of terminology used in naming wing bands. Terms used herein Band h Band sc Band r-m Band dm-cu Apical band Bush (1966) basal medial intercalary subapical apical Steyskal (1979) subbasal discal — preapical apical Foote (1981) subbasal discal accessory subapical apical costal; discal White (1988) subbasal discal — preapical apical White and subbasal discal accessory preapical; apical; EIson-Harris , costal; V band, S band, ( 1 9 9 2) S band, in part V band, in part in part Foote et al. subbasal discal; intercalary subapical; apical; ( 1 9 9 3) costal, V band, S band, in part in part V band, in part Merz (1994) subbasal discal accessory preapical apical 9 3 Table 3. Classification of genera included in this study (after Foote et al., 1993). Tephnfidae Dacinae Dacini C e r atiti n i Trypetinae Euphrantini Epochra Toxotrypanini T ry p e ti n i Carpomyina Cerpomya Cryptodacus Goniglossum Heywardine Myioperdelis Rhegoletis Rhegoletotrypeta Zonosemata Trypetina Acidia Euleia Streuzie Trypeta Unplaced Chetostoma Myoleje Oedicarena Pare terellia 94 Table 4. Distribution and larval hosts of specimens examined. §pecies Acidia cognate Cerpomya incomplete C. schineri C. vesuviene Ch etastama californicum Ch. curvinerve Ch. rubidium Cryptodacus tau Epochra canadensis Euleia fratria Eu. heracleii Eu. uncinete Goniglossum wiedemenni Heywardine cuculi H. cuculiformis Myioperdelis pardalina Myoleje Iimata My. lucida My. nigricornis Oedicarena beameri O. latifrans O. nigra O. persuase O. tetanops Paraterellia immaculate P. superba P. varipennis P. ypsilon Rhegoletis acuticarnis R. adusta . almatensis . alternate . a. orientalis . besiole bateva berberidis berberis blanchardi boycei caucasica cerasi chionanthi cingulete complete . converse . carnivore . ebbettsi . electromarpha mmmmmmwmmwmemmmmmm Distribution Larval host Palearctic Palearctic Palearctic Palearctic Nearctic Palearctic Nearctic Neotropical Nearctic Nearctic Palearctic Nearctic Palearctic Neotropical Neotropical Palearctic Nearctic Palearctic Nearctic Nearctic Nearctic Nearctic Nearctic Nearctic Nearctic Nearctic Nearctic Nearctic Nearctic Neotropical Palearctic Palearctic Palearctic Nearctic Palearctic Palearctic Nearctic Neotropical Nearctic Palearctic Palearctic Nearctic Nearctic Nearctic Neotropical Nearctic Nearctic Nearctic Tussi/aga farfare Zizyphus spine Rosa glutinosa, R. pulverulente not given not given not given not given not given not given Angelica sp. not given not given not given Salanum trichoneuron not given C. [=Curcubita?] mela var. nuski, melons, gurken (cucumber) llex cassine, I. opace, l. vamitaria not given not given not given not given not given not given not given Juniperus deppiane not given not given not given not given not given not given Rose rugose not given Rose sp., R. blanda, R. acicularis not given Berberis vulgaris Berberis equifalium, B. nervosa not given Juglans sp. not given Lonicera xylosteum Chionanthus virginicus Prunus serotine, domestic cherry Jug/ans regia, J. hirsute, Persian walnut Salanum tomotillo Camus amomum, C. a. amomum, C. faemine not given Camus foemina, C. racemasa Table 4 (cont'd). 95 §pecies Distribution Larval host R. emiliee Palearctic not given R. fausta Nearctic Prunus emarginete, sour cherry R. ferruginea Neotropical Solanum sp. R. flavicincta Palearctic not given R. flavigenuelis Palearctic not given R. “flarida' Nearctic Camus flarida R. indifferens Nearctic Prunus emarginete, domestic cherry R. jamaicensis Neotropical not given R. jug/andis Nearctic not given R. juniperina Nearctic Juniperus viginiana, Juniperus sp. R. kurentsovi Palearctic not given R. lycoperselle Neotropical not given R. macquertii Neotropical not given R. magniterebre Palearctic not given R. meigenii Palearctic not given ' R. mendax Nearctic Vaccinium arbareum (as Betrodendran), V. carymbosum, V. pennsylvanicum, V. stamineum, Vaccinium sp., huckleberry, , lowbush blueberry R. meta/lice Neotropical not given R. mango/ice Palearctic not given R. nave Neotropical So/anum muricatum, S. nigrum R. obsolete Palearctic not given R. osmanthi Nearctic Osmanmus americanus R. penela Neotropical not given R. persimilis Nearctic not given R. pomonelle Nearctic Crataegus mallis, C. maleaides, C. opace, Crataegus sp., sour cherry, hawthorn, apple, wild plum R. psalida Neotropical not given R. ramosae Nearctic Juglans major var. glabrate R. reducte Palearctic not given R. rhytida Neotropical not given R. ribicola Nearctic not given R. saute/late Palearctic not given R. striatella Nearctic Physalis heterophylla, P. longifolia R. suavis Nearctic Juglans nigra, walnut, butternut R. tabelleria Nearctic Camus stolonifera, Vaccinium pervifolium, V. avalifolium R. nr. tabelleria Nearctic buffaloberry (? Shepherdia sp.) R. tomatis Neotropical not given R. turanica Palearctic not given R. zephyria Nearctic Crataegus daug/asii, Symphoricarpas albus, S. albus var. laevigetus, S. rivularis, Symphoricarpas sp., snowberry R. zernyi Palearctic not given R. zoqui Nearctic not given Rhegoletotrypeta ennuiate Nearctic granjeno hausteco (=Celtis pallida [Norrbom 1994]) Table 4 (cont'd). 96 Species Distribution Larval host Rh. pastranai Neotropical not given Rh. rohweri Nearctic not given Rh. uniformis Nearctic Celtis sp. Streuzie intermedie Nearctic Rudbekia sp. (prob. lanciniata) S. Iongipennis Nearctic Helienthus tuberosus S. perfecta Nearctic Ambrosia trifida Trypeta fracture Nearctic not given T. inaequalis Nearctic not given T. torti/e Nearctic not given Zonosemate electe Nearctic Solenum caralinense Z. scutal/eta Neotropical not given 2. vidrapennis Neotropical not given 2. vittigere Nearctic Solenum elaeagnifolium 97 Table 5. Character-state matrix used in cladistic analysis. Charactersa 0 0 0 O O 0 Speciesbvc 00110002000012210000002202000000000000000000000 00000000000000000000000000000000000000000000000 000011100o000o00oooooooooooooooooooooooo0000000 11 1 1 1 111 1 1 01110000000000010000 00000000000000001000000000 11 1 1 1 4| 1 0111000000.0..00oo010000.00000000910nunoo11111111:0..01011. 111.. 111 4| 1| 1| 1 1 1 1| 01-11- rvr10 oor11110 o to I11 r1111 91 1111111111101 :11 000 000 0 0 0 O 0 0 0 1111 4| 11 4| 1 1 4| 0111000000:0..0201111000000110..110r0:00..0011110900v00010101 00000000011100000000000000000000000000000000000 o0000000o11100000000000000000000000000000111100 I m .m a U n a t. a .m a m fa k t I S u n e n m a c r r s n I I e” d a O P O s d r O c m m U" kW 8 W m u a” a new. wwmpa m m w WM! 3 wsmm Mm .wkm I a 5 seem... sais a 3% m 9 rammMuwmiaumfldflaa s m.mensm.mflfl yawnia se a OWMMOWWM ImM“..de.Mm msflnmoflnweeflanmmm. % Wm SW... C fill a c r are eeHer bb e aen .I V fint U co ar tau Iuagr e.lp Otleeran.0.l C unoa OU mewmm MmuMuchwmlwmetmwwwngbwadMmeWmWvamca mm vmnmwmm.mmw hwxw pmmv m i a or” bmz [we r a. u o one one ooooococnoo-ooooaooo AccccccCEEEEGHHMMMOOoOPPPPRRRHRHRHHRHRHRHRHHRHR 98 Table 5 (cont'd). Charactersa 0 0 0 0 0 0 Speciesbuc 00000000000000000000000000000222200002222 1 001 91011104|O4|4I4|4|4|4|4|4|111001111111100000000 0 00000000000000000000000000000000011100000 00000000000000000000000000000000000000000 1 000000000090000000000000000000000011100000 4| 4| 1| 4| 4| 4| 4| . - . - l . - . 0 00 0000100001000000000000011000000000000 0 4| 4| 4| 4| 4| v4lo4|4|4|4l4|4|4| r4|4|4|4|4|4l4|004| v4|4|4|4l4|4|000000004|4|0 1 0 0 0 0 4| 4I4|4|4| 4|4|4I 4|4|4I4|4|4|4|4I4|4|4|4|4|4I4|4|4|1l4|4| r4l4l4|4l4|1|1l4| 11990 9111111 0 0000 000 11 4|4| 4| 4| 11 4| 1| 4| 1| r r00 r r1111 :1 r4|4l4| r 91101: :11 r 91100000000 1| 00 00 0 0 00 0 00 0 00 000000000000000000ooooooooooooooooooooooo 01110011000000000001100000000010000000000 01.1110 1 i111 .. .1114! ..1 ..4I4l4l .111000000001111 0010 00 0 0 0 strla fella tabellarla rohweri 'florida ' Rhegoletotrypeta annuleta Rh. pas trenel Rh. Strauzla In term edle S. langlpennls S. perfecta Trypeta lneequells Zonosemata electe R. magniterebre Z. saute/late R. ferruginea R. melgenl R. adusta R. flavlclncta R. kurentsavl R. mecquartl R. jameicensis R. electromarpha R. perslmllls R. nr. tabelleria R. psallda R. rhytida R. zernyi R. blanchardi R. mongol/ca R. besiole R. batava R. flavigenuelis R. caucasica R. pomonelle R. R. rlblcala Rh. uniformis Z. vidrapennis Z. vittigera R. R. jug/andis R. mendax R. zephyria R. sue vls ' R. 99 Table 5 (cont‘d). Charactersa 2 1 1 1 Speciesbvc 4|0000000011100000000000000000000000000000000000 4| 4| 4|4| 4| 4| 0111001100001111001111000011114I4l4l4l4l4l4l4l4l4l4l4l4l4l4l4l1 011100010000111100000000001010100111111-11000011 0000000o0000000000111100oooooooooooooooo0000000 4| 4| 4| 4| 4| 11 1114! 4| 4| 4| 0 O I. I. I. I ’ " "" 9 100 ’0 I. 111 00000000001000000000000000001111111 O 4...o 00000000000000000001000000000000000000000000010 01110000000000010000000oooooooooooooooooooooooo 001100000000100100000000ooooooooooooooooooo0000 00000004lnu00004|000000000000000000000000000000110 10000000011100100000000000000000000000000000000 m m m u n a a m .m a m m m e n m a m c r I w b e” d a 0 am M wmwa wwmpm .m m w MM! 3 wsmm mm wbm I a as aaamc. sais a aalm m 9 wammMummiaumflJaflaa s ”.mansmmnfl .mmmniu s“ o wnnmownu rmmxwmd.mMammammmumwumammmmmm a my swa ii 0 r cr ale .1 V ...unt vaflbudm anorcaa rsmrm. sle.m.rrmtgnwua em... 0. ars hocwomcMhmwuwmeQMme merPonnmmememmmeumwmmow mmswa .wwmh my ma.wnpammsywam cacmmmcmmMcwmmcr caOOthh puumoalyyyelOtlallflhfltflll0.0.0.0..-CO... ACCCCCCCEEEEGHHMMMOO00PPPPRHRHRHHRHRHRHRHRHRHRH mo Table 5 (cont'd). Charactersa 2 1 1 1 §p_eciesb-° 00000000000000000000000000000000011100000 4| 4| 4| 4| 4|0 104|4|4|1|4l4|4|0 14|4|4|4|4|4|4|4|4|4| 111111010 100000000 0 O O 0 4l4l4l4l4l4l4l4l4l4l4l4l4l4l4l4l4l4l4l4|4l4l4l4l4l4l4l4l4l4l4l4l4l0004l4l4l4l4l 11 00000000000000000oooooooooooooooooooooooo 4|4|4|4|4|4|4|4|4|4|4|4|4|4|4|4|4|4|4I4| 14|4|4|4|4|4|4|4I 1O 1 1000 1111111 0 0 00 0 4| 11 4| 4| 100011.0011100101111101.001011010110000000001110 0000000000000111100000000000000000.0000000 00000000000000000000000000000000000000000 00000000000000000001100000000000000000000 1 1 nunuo0004I4l000000141.14.-000nuo4l0nu00001141100001100.J 00000000000000000001100000000000000004l4l4l4l m m U m a a I w mm a 8 ac m a 4 ms Wk 8 8.! SH h I 9.! e e iawmxm m m ur m mm “mmnsmn u s n dnsasell d Im a! re ataeiln hm m 8.! rltununl n. aMIa [9.58m1U0rwm e .I Mna mmmMWmmmngmamMaymwummmWnemnwmuwmmwaamupe www“nu.mmeommammmmtumeMWwwuwaprmwmmaummm wrwurkmj mmmmp7mm mpmr smtasznn.mlmmewW . o a o a n c o o 4 o o o n a c a o o o u o o 4 a a a a . ht a 4 O a - RRRRRRRRRRRRRRRRRRRRRRRRRRHRRRRRRSSSTZZZZ 101 Table 5 (cont'd). 3 Charactersa 2 3 3 2 2 Speciesbvc 1 0000000000001000000000000000m010011111111111110 1 1 1 11 1 00000000000000000ooooooooomoooooooooooo00010100 00000000000000000100000000000000000000000000000 00000000000000001000000000000000000000000000000 10004!110011110oo1111111111100101100000000000001 0000000000000000000000000000090000000010110004110 000000010000000000110000001011111nw010..0.110:111111. 00000000000000000000000000000000011110000000000 000000o000O00000000ooooooooooooooooooooooo00010 m .m a u n m M s M... n m a s c r I r .1: 9! d n a a P ..m 5 dm r O c m H an be a .U m H e 3P H m u 0 8 ea WU, a n r a” r a I a I a 3 MI... a 98"... mmaomm a 8.8 “3.08 Sowis 8 all: r a. wammMummiaunHauaa s ”mansahflfl yawnia sw o oanoMmu rmMuwmd.mMam unmwummmmummmmmm m aw swa 0 [iii a C f Or 69,9 6" .I V ”.04.. raocwomcmhMeuwmeWMcMw anPounueemmmdmonuowmmow mmsmu .men my wo.mnpammsywam bcmcMchmmMcwmmcr caoohlnohrPUUoUIoauyyyecltuaO.lhcoOCIoIcc-0000000000 ACCCCCCCEEEEGHHMMMOOOOPPPPRRHRRRRRRHHRRRHRHRHRR 102 Table 5 (cont'd). Characters8 3 3 3 2 §peciesbv° 1 0007.00000oooooooooooooooomooooooo00000000 4|4|4|4| 90111010111.1111111101111111110000 000 o. O 0 0 4| 4| 4| 0011 1011101011111111111 t04|4|4|4l4|1|4| 100000000 0 O 0 00000000000000000000000000000000000000000 01110000oooooooooooooooooooo0000000000000 00000000000000000000000000000000000000000 11 10000000010000000110010001100000011110000 1 1 00000000000000000000000000000000000000000 O00000000000011110ooooooooooooooooooooooo m a I u m a a M ”88 a m m 84m a. a! .w sa .M 9! wk WM 9 .awmsm a m an a mm klestn m s n .mnsawelk I [m a” to a wiMn a .m a! rltunnnl e. amla la .3... ludm m e .mtan .mg 30" eneo na .10le atslmumynttwr pa atem mmamWemmgggmaoMWmeummMWnemnWeshmaWaamkpe Cfflnawaaaflwbflmw hetaWbU—dhberNmmmflnetemwmm} whuhflkmmmmmxmmflmWPLWhfiswwmwua9 uuoflememm 0 .mb .1 bb . mz n r .l e fl..nah.h.h.mIPV.Msvv . o o o n o o o o o o o o o o a o o o a a o o n o u o o t a or o o o HRRHRRRHRRRRHHHHHRHRHRRRHRHRRRRRHSSSTZZZZ 103 Table 5 (cont'd). Charactersa 4 4 4 4 4 Speciesbvc 00000000000000000000000000000000011111111000000 01111111011111110010000001111111111111111111111 11 0000000nu000000000000000000000000011111111000001 01110O10000011114!1111111111111111111114l4l4l4l4l4l4l4l4l 1000000000000000000000111410000000000000110100011 00000000000000000000000000111010011111111000010 0114i29.01000011110000000000111111111111111111111 00000000000000000000000000000000000001111000000 00000000000000000011111111000000000000000000000 00000000011100000011001111000000000000000000000 m .m m u n a a a .w a .m I .w t n m I s u n e r s a n c r I 0 I e” d O a O p (I s d" r r m .m am a wmwa wwmpm m .m w mm” ammwmw mmimmm a a aka awamm swim m aw m wnmmmmWchmm3mwuqmm msmmmmmmmwuummmem 9 mm swa c lit. r a cc r are eeflerebb e aen .Ia Ve fint hacwomcbhmeuflumwpmmnvm.” mwupuouunuma.mmmdmonuowwmow wmsma .Wmmh my wo.mnpmmvsyaaw cmcmmmcmmMcwmmcr ca..hhh Puuuaa.YYYe..Ja...h.................... AccccccCEEEEGHHMMMOoOOPPPPRRRRRRRRHRRRRRRRRRHRR 104 Table 5 (cont'd). Charactersa 444 4 Speciesbvc 00000000000000000000000110000000000000000 1000111111111111111111.-1111111111100001111 0009.4|000000000000000000001100000010110100 1 1 1000010011010111.0..1100000o11100011000000000 1100110000.00010..0..0..110010..10..0..101011000000000 0000010011111000o110010111111000000000000 11111111111111111111111111111111100001111 oooooooooooooooooo00000110000000000000000 00000000000000000000000000000000000000000 1 011000000000000000000101000000000000000000 00000000000000000000010001100000000000000 m ra u m a a I w mm m a m ms w“ a a! sh h I 9.! a.“ e iuwmbm u m H, m mm kansmn m m n dnsaSLOI-II e [m a, ’0 atae.l.'n "m n a! rItUNInI n. Mla ’9 .SHMIuor m 9 cl “"3 amQWWmmmgmMmammaymwummmWnemnwmnwmmwaamupe cwsnauaaaeomwommMetaWbuamucu.woprnznmtmmm.m wfwmflkmmmmmmeWeePLWhflsttmzmg uubeWome C ab i1 .Db m2 n r .l e th.h.h.m pYMSVV I I I I I I I I I I I I I I I I I I I I I I I I I I I I t I I, I I I HRHRRRRRRRRHRRHHHRHRHRRRHRHRRRRRHSSSTZZZZ 105 Table 5 (cont'd). Charactersa 555 5 §Qeciesbvc 00000001000001000001111111000000000000000000000 11111111011111111101111101111111111111111111111 00000000000000000011111111000000000000000000000 00000000000000001111111111000000000000000000000 10001110011100001100000000000000000000000000000 000000000000o0000011111111000000000000000000000 10000000000000000000000000000000000000000000000 01110001000010010000000000111111111111111111111 00001110000000000000000000000000000000000000000 21111111011111111111111111111111111111111111111 o1110001000011110100000000111111111111111111111 01110000000000000000000000000000000000000000000 m .m m u n a a s m m a m s m m e n m a c r I f .6 0.! d n a O P (O: 8 dm .1 m m .m m ” un mesa ” II mm a mafia wwmpm m .m w am usnn am rsm a m a h a s 5 .IS a a, r m nammmUaaiaummwuaa “ MMansmmnn ammni“ se .0 ofimmaflnu ammumma.mMawmammmummmuammmmmm m am sWa c .I.It..l a c f Or 819 n .l V .0"... MhVSbUdr mnorcaa 8fSW’hIPPSI9b-ffflflgnflameSinp ars .mocwwmcmmweumymeWMMMW wquWMnmmememmoWoWOwmmww mmsme .wOmh ny .mw.dnpm_r.vsya .w cmcmmmcmmzcwmm r ca I IhlnIh puu-moa I ya I I I3 I I Ih I I I I I I I I I I I I I I I I I I I I ACCCCCCCEEEEGHHMMMOOOOPPPPRRRRRRHRRHRHRHRRHHHRR 106 Table 5 (cont'd). Charactersa 5 5 5 5 Speciesbvc 00000000000000000000000000000000000000000 III!111-111111111111111111111111111111111111 000000oooooooooooooooo0000000000011100000 00000200o200000o0000000000000202200010000 00000000000000000000000000000000011110000 00000000oooooooooooooooooooo0000000000000 00000000000000000000000000000000000010000 11111111111111111111111111111111100000000 00000000000000000000000000000000000000000 4|1111111111111111111111111111111111120000 11.111111111111111111111111111111111111111 0000000000000000000000ooooooooooooooooooo m h u m a a m “a a a e M»... A” 8 MM ms W.“ 3 al Sm h I 6! e e lawman “ m m, m mm mansmn n a m dnsasMIl e m a! ro .Uataeiln nm n 89 ritunlnd nu BM/a I8 ..bMMI Ome 9 [34603 .n acna.ene a .rledaotsdl Yntt r Pa aer smmmecmgmMmammwymwuummWnemnwmmwmmwaamnpg mrwnauaaaeobw wnmetaWbUahflCflWOPrnznmtflQMI c ab .1 bb mz n r .l e f.hah.h.h....rm PWMSVV RRRRRRRRRRRHHRRRRRRRRRRRHRRRRRRRRsaarzzzz 107 Table 5 (cont‘d). Charactersa 6 6 6 Speciesbvc 000o1110011000000000111111000000000000000000000 00001110000000000100000000000000000000000000000 00oo1111000000001100000000000000000000000000000 00001110000000001100000000000000000000000000000 0111.111100000111111111111111 91011 v 1011111111100 0 00 01111110000010010000110000000000000000000000000 .1.1111111011111111111001111111111111111111111111 011100010000011000009.20000011000000000000111100 11111i..l11011111111111111111111100000000000000001 01110001000001100022220000011010011110000111111 00O00000000000000000000011000024.433333333222220 00000000000000000000000000000000000000000000000 m .m m U n a m m m .m. s m m w m .m e” .M m u w m. ..n m .mu m n m m I "9 Icsp I m t MM 0 rtam cm a I u m a a ml! ammsnt MmaOMm a 3.3 Mews .sMis m a” r a. wamuluwmaaunflaflaa s HnMnsaMrfi maMmie mw W mnnmawnu ramuwmm.mMammumrmumwmmummmwmm w ea 3mm .1. e i v fl mMVflMudm .ranorcaacflrsmr.m.w..mle.m.rrmmgnwmamesino. ars .mocwomchhbeuwmeWMcMw wwupwnnmmemedmoWouowmmow mmswa .w”Mh my mo.mnpamvsyaam cmcmmxcmmmcwmmcr ca..hhhrPuumoa.YYYe..ya...h..u................. AccccccCEEEEGHHMMMOOOOPPPPRRRRRRHRHRRRRHRRHRRRR 108 Table 5 (cont'd). Charactersa 6 6 Speciesbvc 00000000000000000000000000000001101110000 00000000000000000000000000000100000000000 00000000000oooooooooooooooooooooo00000000 O 0 0 00000000100000000000000000000000000001111 11111111111111111111111111111111111111111 0111011100000000o001101000000101100001111 1 011111110111100001100010000111111111110000 111101111111100o0001101001111101100001111 00000000000002222002220220000020000002222 00000000000000000000000000000000000001111 M II u m a I a aI .m a m ms WM 3 She h I e! e e iawmem m M Mr m mm ImmnsMn u I .m dnsasmll 9 Im a! re ataeiIn nm n 89 rIIunInI n. aMIa I9 .fiIIquoer 9 Iamna .m acheneo a .nledaotsdl Yntt r Pa aer amImecmngmammmymwquIWnemnwmummmwaamnme wawmhMmmmeamwowMMmzmwmmeuxamwprmummaummw I If .t I . I II .II m Mb k mmMMPme um sW ezflMhhhmlWmevv I I I I I I I I I I I I I I I I I I I I I I I I I I I I I t I I I I I RRHHRRRRRRRHHRRHRRRRHRRRRRRRHRRRHSSSTZZZZ Table 5 (cont'd). 109 Speciesb-c Acidia cognata Carpomya Incompleta C. schineri C. vesuviana Chetostoma californicum Ch. rubidium Ch. curvinerve Cryp todacus tau Epochra canadensis E ulela fratrla Eu. heraclei Eu. unclnata , Goniglossum wiedemannl Heywardine cuculi H. cuculifor‘mis Myioperdelis pardalina Myolela .Ilmata My. lucida Oedicarena lat/Irons O. nigra O. persuasa O. tetanops Paraterellia Immaculata P. varipennis P. superba P. ypsilon Rhegoletis acuticarnis R. alternata R. juniperina R. berberidis R. berberis R. cerasl R. almatensis R. clngulata R. chionanthi H. indifferens R. osmanthi R. completa R. boycei Fl. ramosae R. zoqui R. converse Fl. chopersella Fl. nova R. tomatis R. cornivora R. fausta AddO—AO—L—L—L—L-A-L—l—L—l—L—ld‘ddOOCO—I—IOOO—L—tooooooooooo-L—I—toQ\l Charactersa 7 OOOOOO0°COOOOOOOOOOOQOOOOOOOOOOOOOOOOOOOOOOOOOC01 7 ocoooococooooocococococooocoo‘ooccoccoocdddocoom OOOOOOGOOOOOOOOOO-‘OOO-‘dd-‘OOOOOOOOOOOOOOOOOO-‘d-‘ONN Table 5 (cont'd). Rhagoletotrypeta annulata Rh. pastranai Rh. rohweri Rh. uniformis Strauzla Intermedia S. Iongipennis S. perfecta Trypeta lnaequalls Zonosemata electa Z. scutellata Z. vidrapennis Z. vittigera Charactersa 7 7 7 7 7 fleciesbvc 3 4 5 6 7 R. caucasica O 0 0 O 0 R. ferruginea 0,1 0 0 0 0 R. adusta 1 ? O O 0 R. blanchardi O 0 0 0 0 R. flavicincta 1 1 0 0 0 R. kurentsovl 1 0 0 0 1 R. macquarti 1 0 0 0 O R. jamaicensis 1 0 O O 0 R. magniterebra O 0 O O 1 R. melgenl 1 0 0 O 1 R. mongollca 1 0 0 O 0 R. basiola 1 0 0 0 O R. batava 1 0 0 0 0 R. pomonella 1 0,1 0 O 0 R. 'florida“ 1 0,1 0 0 0 R. mendax 1 0 0 0 R. zephyria 1 0,1 0 0 0 R. persimilis 1 1 0 ‘0 0 R. nr. tabellaria 1 0,1 0 0 0 R. psalida 1 0 0 0 R. rhytida 1 O O 0 R. ribicola 1 0 0 0 R. striatella 1 0 0 0 R. suavls 1 1 0 0 0 R. jug/andis 1 1 0 O 0 R. tabellarla 1 0 0 0 R. electromarpha 1 0 0 0 R. zernyi 1 0 0 1 R. flavigenuaIis 1 0 O 1 0 0 1 0 0 0 0 0 1 0 1 0 1 0 1 0 0 0 0 0 0 0 0 0 0 0 O 0 0 0 0 0 0 1 0 O 0 1 0 0 0 1 0 0 0 1 0 0 OD COCO-*OOOOOOOOOOOI I OO-)O aMonomorphic characters and their states appear in bold typeface, and polymorphic characters and their states appear in plain typeface. bA species in plain typeface is redundant in monomorphic characters with the species in bold typeface immediately preceding it. °Ch. rubidium is redundant with Ch. californicum for all characters. 111 Table 6. Characters occurring in single species. Head Median occipital sclerite with a shelf-like protuberance: Chetostoma curvinerve Proboscis geniculate: Goniglossum wiedemenni Labellum with capitate setae: Paraterellia ypsilon Face with a pair of dark spots: Clyptodacus tau Flagellum with an apical fringe of black setae: Cryptodacus tau Thorax Anatergite with long, fine, erect hairs: Epochra canadensis Presutural supra-alar seta absent: Epochra canadensis Postsutural acrostichal seta absent: Oedicarena nigra Katepistemal seta absent: Acidia cognam Wing Whitish spot at apex of cell r4+5: Epochra canadensis Abdomen lnvaginated sac-like structure in pleura between segments 4 and 5: Myoleje Iimata Tergal setae of female grading from relatively long medial ones to shorter lateral ones: Epochra canadensis Male Genitalia Basiphallus with a pair of small tubercles ventrally at its base: Paraterellia immaculata 112 Table 7. Characters used in cladistic analysis. 10. 11. 12. 13. 14. 15. 16. Flagellum rounded or angular dorsoapically and without a detectable point (0); or flagellum more or less angular dorsoapically and with at least a small dorsoapical point (1). Distal half of arista bare or with a few scattered microtrichia (1); or arista uniformly microtrichiose (O). Facial ridge about as wide as or narrower than parafacial (0); or facial ridge decidedly wider than parafacial (1). Genal seta concolorous (0) or not concolorous (1) to principle head setae (excluding gular, postocellar. and postocular). Gular seta concolorous (O) or not concolorous (1) to principle head setae (excluding genal, postocellar, and postocular). Postocellar seta concolorous (0) or not concolorous (1) to principle head setae (excluding genal, gular and postocular). Postocular seta concolorous (0) or not concolorous (1) to principle head setae (excluding genal, gular and postocellar). Upper orbital seta absent (1) or present (0). Genal setae enlarged, numerous, or both (1); or genal setae not enlarged or unusually numerous (0). Male with frontal setae pointed and similar in size to the frontal setae of female (0); or frontal setae of male blunt and larger than frontal setae of female (1). Ground color of scutum yellowish (0); or ground color black or brownish (1). lntegument of scutum with a Carpomya-Iike pattern (1); with a whitish or yellowish medial stripe or prescutellar spot (2); or more or less uniformly pigmented or with an intraspecifically variable pattern (0). Disc of scutum with microtrichia (0); or disc lacking microtrichia, scutum with peripheral microtrichia only (1). Disc of scutum with setulae of uniform color (0); or disc of scutum with a mixture of light and dark setulae (1). Supra-alar area with unmodified microtrichia (0); or supra-alar area with black, velvety microtrichia (1). Mediotergite with simple microtrichia (O); or mediotergite with pollenose microtrichia (1). Table 7 (cont'd). 17. 18 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. Halter wholly yellowish or brownish (0); or halter with the stem yellowish and the knob dark brown or black (1). . Outer scapular seta concolorous (0) or not concolorous (1) to principle thoracic setae (excluding presutural acrostichal, and proepisternal setae). Proepisternal setae concolorous (0) or not concolorous (1) to principle thoracic setae (excluding outer scapular, and presutural acrostichal setae). Bare spots at the inner ends of the transverse suture and base of postsutural dorsocentral seta (1); or transverse suture and base of postsutural dorsocentral seta without bare spots (0). Band r-m present (0) or band r-m absent (1). Band so not crossing vein r-m (0) or band sc crossing vein r-m (1). At least proximal hairs of fringe of upper calypter dark brownish or black (0); or all hairs of upper calyptral fringe whitish (1). Cell br within band sc with a hyaline spot (1); or cell br within band so entirely pigmented or part of a larger hyaline area (0). Wing pattern with bands so, r-m, and dm-cu fused anteriorly, and bands h and sc fused posteriorly (1); or wing pattern with one or more of these bands not fused as described (0). Wing pattern with bands h, so, and dm-cu free posteriorly, band r-m absent, and the apical band with the posterodistal comer of the anterior arm well ahead of vein M (1); or wing pattern otherwise (0). Hind femur wholly yellowish (0) or infuscated (1). Tarsomere 4 or 5 or both same color as rest of tarsus (usually yellowish) (O); or darker than basal segments (1). Mid tibia with a distinct posterodorsal row of setae (O); or mid tibia without a distinct posterodorsal row of setae (1). Hind tibia with a distinct anterodorsal row of setae (O); or hind tibia without a distinct anterodorsal row of setae (1). Mid femur or hind femur or both with enlarged setae ventrally (1); or both femora with setae not enlarged (0). Males with anteroventral row of setae on fore femur enlarged (1); or anteroventral row with setae on fore femur normal, not enlarged (O). Table 7 (cont'd). 33. 34. 35. 36. 37. 38. 39. 40. 41. 42. 43. 44. 45. 46. 47. Fifth tarsomere relatively small, cylindrical, about twice as long as maximum diameter (1); or fifth tarsomere larger, flattened, less than twice as long as maximum diameter (0). Ground color of terga yellowish (0) or brownish to black (1). Excluding tergite 1, one or more terga with bands of light and dark colored setae (1); or setal color of terga uniform (0). Sternum 7 of male with polygonal sculpturing (1); or sternum 7 of male without sculpturing (O). Basiphallic vesica present (1) or basiphallic vesica absent (0). Ejaculatory apo‘deme with distal edge flared (1); or ejaculatory apodeme with edge coplanar with blade of apodeme (0). Dorsal portion of epandrium produced posteriorly well beyond base of surstyli, the angle formed by posterior edge of epandrium below proctiger and long axis of surstyli decidedly less than 90° (1); or dorsal portion of epandrium not markedly produced posteriorly, the angle formed by posterior edge of epandrium below proctiger and long axis of surstyli about 90° or more (0). Hypandrial sac lined with numerous heavily sclerotized denticles (1); or hypandrial sac not lined with denticles, or intrahypandrial membrane not forming a sac (0). Sub apical distiphallic lobe trumpet-shaped (0); an elongate lobe or flap (1); or with a pair of large apical hooks (2). Bacilliform sclerites with a dorsal keel, at least distally (1); or bacilliform sclerites rounded dorsally and without a keel (O). Microtrichia present on base of surstyli anteriorly (1); or base of surstyli bare anteriorly (0). Membrane connecting bacilliform sclerites to surstylus with microtrichia present (0); or membrane connecting bacilliform sclerites to surstylus bare (1). Epandrium with. numerous, evenly distributed microtrichia (1); or epandrium without microtrichia or at most with a few patchy ones (0). Syntergosternum7+3 with one or more macrochaetae (0); or syntergosternum7+3 with only microtrichia or bare (1). Parameral sheath of distiphallus with polygonal sculpturing (0); or parameral sheath of distiphallus without polygonal sculpturing (1). Table 7 (cont'd). 48. 49. 50. 51. 52. 53. 54. 55. 56. I 57. 53. 59. 60. 61. 62. 63. 64. 65. Tips of surstyli with a cluster of long setae (1); or tips of outer surstyli with setae shorter and not forming a cluster (0). Surstyli with Carpomya-Iike setae distally (1); or surstyli with normal setae (0) Hypandrial apodeme present (0); hypandrial apodeme absent (1). Surstyli with anterior lobe only (0); surstyli with anterior and posterior lobes (1); or surstyli with anterior, medial, and posterior lobes (2). Right pregonite deflected ventrally (O); or right and left pregonites even (1). Acrophallus present (1); or acrophallus absent (0). Inner prensiseta on a large tubercle that places it decidedly distal of the outer prensiseta (1); or inner and outer prensisetae at about the same level (0). Hypoproct entire (0); or hypoproct divided (1). Hypoproct extending dorsally for most or all of the height of the proctiger (1); or hypoproct extending dorsally for less than half the height of the proctiger, if at all (0)- lnner and outer prensisetae similar in size (0); inner prensisetae larger than outer prensisetae (1); or inner prensisetae smaller than outer prensisetae (2). Anterolateral corner of bacilliform sclerites forming lobes (0); or anterolateral corner of bacilliform sclerites not forming lobes (1). Apex of aedeagus enclosed by parameral sheath (1); or distal portion aedeagus not enclosed by parameral sheath (0). Basiphallus with membranous ventral keels (1); or basiphallus without membranous ventral keels (O). Vesica contiguous with phallotheca (1); or vesica free distally (0). Subapical distiphallic lobe bare (0); with numerous sclerotized denticles (1); microtrichiose (2); fimbriate without supernumerary lobe (3); or fimbriate with supernumerary lobe (4). Total number of spermathecae three (0); total number of spermathecae two (1); or total number of spermathecae four (2). Spermathecae cylindrical (0); or spermathecae globular (1). Number ofspermathecal ducts: 3 (O); 2 (1); or '4 (2). Table 7 (cont'd). 66. 67. 68. 69. 70. 71. 72. 73. 74. 75. 76. 77. Eversible ovipositor sheath about as long as segment 8 (1); or eversible ovipositor sheath distinctly longer than segment 8 (0). Eversible ovipositor sheath with microtrichia proximally (1); or eversible ovipositor sheath without microtrichia (0). Denticles on eversible ovipositor sheath near segment 8 with single point (0); or teeth near segment 8 with multiple points (1). Large discal denticles on ventral surface of eversible ovipositor sheath triangular and with a single point (0); or large discal denticles on ventral surface of eversible ovipositor sheath squarish and irregular apically (1). Segment 8 constricted at base (1); or segment 8 not constricted basally (0). Segment 8 with tip laterally flattened (1); or segment 8 with tip dorsoventrally flattened (0). Segment 8 with microtrichia or denticles or both around cloaca (0); or cloaca glabrous (1). Tip of segment 8 with subapical points, projections or serrations (0); or tip of segment 8 with single, apical point (1). One spermatheca definitely smaller than other(s) (1); or spermathecae nearly the same size (0). Spermathecal ducts definitely annulated and radiator hose-like (1): or spermathecal ducts smooth (0). Dorsal taeniae extend to segment 8 (1); or dorsal taeniae not reaching segment 8 (0)- Ventrally, tip of syntergosternum 7 with about 8-16 stout setae (1); or setae at tip of synterflsternum 7 with setae of normal size (0). 117 Table 8. Characters not included in the cladistic analysis. Character Variation Head shape of flagellum continuous color of principle setae ambiguous color of genal and gular setae ambiguous color of postocellar seta ambiguous size of ocellar setae continuous size of postocular setae continuous shape of median occipital sclerite discrete size of paravertical seta continuous position of inner vertical seta continuous width of palps continuous size of genal and gular setae continuous size of head setae continuous coloration of gena ambiguous shape of arista ambiguous number of frontal setae continuous number of orbital setae ambiguous attitude of upper orbital seta discrete Thorax ' color of principle setae ambiguous color of scapular setae ambiguous size of scapular setae ambiguous coloration of scutum ambiguous color of katepistemum ambiguous color of postpronotal lobe ambiguous color of setulae on postpronotal lobe ambiguous presence of microtrichiose stripes on scutum discrete coloration of scutellum ambiguous presence of velvety microtrichia on scutum ambiguous coloration of mediotergite ambiguous position of dorsocentral seta continuous presence of dark flecks in cuticle ambiguous size of proepisternal setae continuous number of anepistemal setae continuous number of katepisternal setae discrete color of inner scapular seta ambiguous color of outer scapular seta ambiguous color of, katepisternal seta discrete presence of pleural stripe ambiguous presence of minute setae on halter ambiguous Wing extent of band r-m ambiguous position of band so ambiguous number of apical bands ambiguous coloration of bands ambiguous number of setae on R4+5 continuous presence of spurious vein or bullule in cell r1 ambiguous position of vein r-m continuous 1 1 8 Table 8 (cont'd). Character Variation Wing (cont'd.) shape of lower calypter continuous extent of microtrichia on wing membrane ambiguous apex of wing hyaline or infuscate discrete coloration of wing base ambiguous prominence of proximal and distal subcostal bands ambiguous number of setae at tip of vein R1 ventrally ambiguous Legs coloration of legs ambiguous Abdomen color of tergal setae ambiguous presence of tergal microtrichia ambiguous tergal coloration ambiguous color of setae on tergite 1 ambiguous Male Genitalia bleb present between sterna 6 and 7 ambiguous size of sterna 6 and 7 invariable shape of sternum 5 ambiguous shape of sternum 7 ambiguous length of microtrichia in membrane between sterna 6 and 7 invariable position of mechanoreceptors on sterna 6 and 7 ambiguous shape of phallapodeme ' ambiguous length of distiphallus ' continuous coloration of ejaculatory apodeme ambiguous shape of epandrial phragma continuous shape of surstylus continuous shape of hypandrium continuous size of proctiger continuous shape of hypoproct continuous appressed flap of distiphallus with distal microtrichia invariable size of denticles on subepandrial membrane at base of phallus continuous presence of subapical distiphallic lobe invariable basiphallus with sculpturing ambiguous shape of dorsal keel of bacilliform sclerite continuous shape of prensisetae continuous extent of denticles on anterior surstylar lobe continuous bacilliform sclerite with anteroventral lobe ambiguous vestiture of surstylus ambiguous vestiture of epandrium continuous size of apical distiphallic lobe continuous vestiture of apical distiphallic lobe ambiguous position of prensisetae ambiguous sclerotization of subapical distiphallic lobe ambiguous sculpturing of parameral sheath of distiphallus ambiguous coloration of epandrium ambiguous location of denticles on surstylus continuous number of sensilla on anterior surstylar lobe continuous position of anterior surstylar lobe continuous 119 Table 8 (cont'd). segment 8 twisted Character Variation Male Genitalia (cont'd.) number of gonopores ambiguous shape of apical distiphallic lobe continuous absence of epiphallic sclerite invariable shape of anterior bridge of bacilliform sclerite ambiguous shape of rectal lining ambiguous presence of free sclerite on basiphallus distally ambiguous attachment of subepandrial sclerite to bacilliform sclerites ambiguous position of subepandrial sclerite ambiguous fusion of bacilliform sclerites to surstyli invariable presence of muscle between bacilliform sclerite and epandrium invariable shape of sclerotized portion of parameral sheath of distiphallus ambiguous presence of sensilla on distiphallus ambiguous Female Genitalia - number of spermathecae per side ambiguous color of spermathecae continuous shape of spermathecae continuous ornamentation of spermathecae continuous orientation of spermathecal ornamentation ambiguous length of spermathecal ducts continuous sclerotization of syntergosternum 7 ambiguous shape of small denticles on eversible ovipositor sheath continuous shape of large denticles on eversible ovipositor sheath invariable presence of denticles on segment 8 ambiguous presence of sac-like structure within base of segment 8 ambiguous presence of lateral groove in tip of segment 8 invariable number of sensilla in lateral groove of segment 8 continuous shape of tip of segment 8 ambiguous spermathecae with atrium ambiguous spermathecae with minute capitate structures invariable sterna 10 visible within eversible ovipositor sheath ambiguous ambiguous 120 Table 9. Leg coloration by segment, excluding tarsi. Fore Leg Mid Leg Hind Leg Species Acidia cognata Cerpomya incompleta C. schineri C. vesuviana Ch etostoma californicum Ch. curvinerve Ch. rubidium Cryptodacus tau Epochra canadensis Euleia fratria Eu. heraclei Eu. uncinata Goniglossum wiedemenni Haywardina cuculi H. cuculiformis Myiopardalis pardalina Myoleja Iimata My. lucida My. nigricornis Oedicarena beameri O. Iatifrons O. nigra O. persuasa O. tetanops Paraterellia immaculata P. superba P. varipennis P. ypsilon Rhagoletis acuticarnis R. adusta . almatensis . alternate . a. orientalis . basiola batava . berberidis . berberis blanchardi boycei caucasica cerasi . chionanthi cingulata comp/eta conversa . cornivora . ebbettsi . electromomha mumpmmmpmmmmmmmmmm cxatr fm l+l+l++|+ll ++++I+- I I I+|+I I I+-o' |+l+‘ O+l+l++|+l +-)l++l+ I Il|+ll .¢II+I l|+ll I + I + I I ++++l+l+'+'+'+++' l|+lll+l+ll l+°~)'l+' cx trfm +-o++I+I+-+'+'+++' l||+ll I+-o'I-I-' 'H-H- '~ I + I + I I ++++l+l+'+'+'+++' ll+|+l + +.¢I+.+I cx trfm I + I + I +-o++I+I+-+-+-+++' l '4. l I Il+l+ll 'l+'+++l+' l+' I+-o'+ Table 9 (cont‘d). 121 Species R. emiliae R. fausta . ferruginea . flavicincta . flavigenualis . 'florida“ . indifferens jamaicensis . jug/andis . juniperina . kurentsovi . chopersella . macquarti . magniterebre . meigeni R. mendax R. metallica R. mongolica R. nova R. obsoleta R. osmanthi R. penela R. persimilis R. pomonella R. psalida R. ramosae . reducta . rhytida . ribicola . scutellata . striatella suavis . tabs/[aria . nr. tabellaria . tomatis . turanica . zephyria . zernyi R. zoqui Rhagoletotrypeta annuia ta Rh. pastranai Rh. roh weri Rh. uniformis Strauzia in term edia S. Iongipennis S. perfecta Trypeta fractura T. inaequalis mmrumrarnmgornmzomm IDZDIDIIZDPZDIZDZDD Fore Leg cxa tr fm tb +-+- ‘l+l+l+" C I C "+l+" I I . + I + I .++ l .++ I l+++l+' '+ '+l+++l+' I '+-Jl+++l++l+++'+ '+-\)l+++++l+++ I '+++ I I I I l+'+'+++ I'+. 'l+'+l+'+ I 'l+++l+' I 'l+++ '+++++'+- . + .++ 'l+'+l+++'+++'++l++++++++l++l++++ I.+I .¢I+I Mid Leg cx tr fm '++'+'+++l+l+'+' '+++I+'+'+++'+-oI++++++++I++I++++' U '+ I l '+ I I I|+II IQ! II|+II 'l+++"l+' '+++++'+' . + '++++++++++I++++++'l+++ 'l+l++'++'+++ I '+ U I I I + I '+ I '+l+++'+l+++ 'l+'l+'+l+l+'l+'l++' "++'+'+++I++'+' Hind Leg cx tr frn '+++l+'+'+++'++l++++++++l++++++' I I + '+ I 'l+'l++ I I + ‘+l+'+ 'l+'l+"+'++'l++ 'l+'+ .+++++ 1 22 Table 9 (cont'd). Fore Leg Mid Leg Hind Leg Species cxa tr fm tb cx tr fm tb cx tr fm tb T. tortile - - - Zonosemata electa - - - Z. scutellata - - - Z. vidrapennis - - - Z. viiiqera - - - I |+ I I I I I |+ I I I I I I I I I I I ++u++ aAbbreviations: cx, coxa; tr, trochanter; fm, femur; tb, tibia; -, wholly yellowish; +, infuscate; 1, polymorphic; ?, segment missing. 123 Table 10. Species with tergal patterns matching the medial pattern systema. Species Subfamily Referenceb Abebaiodacus fuscatus (Wiedemann) Dacinae Munro 1984, fig. 98 Acanodacus botianus Munro Dacinae Munro 1984, fig. 105 Acanodacus brevis (Coquillett) Dacinae Munro 1984, fig. 73 Acanodacus ceropegiae Munro Dacinae Munro 1984, fig. 100 Acanodacus cuspidatus Munro Dacinae Munro 1984, figs. 73, 101 Acanodacus serratus Munro Dacinae Munro 1984, figs. 73, 103 Acanodacus viator (Munro) Dacinae Munro 1984, fig. 99 Ancylodacus collarti (Munro) Dacinae Munro 1984, fig. 64 Ancylodacus flavicrus (Graham) Dacinae Munro 1984, fig. 64 Bactrocera albisfrigata (deMeijere) Dacinae White & Elson-Harris 1992, fig. 177 Baofrocera caudata (Fabricius) Dacinae White & Elson-Harris 1992, fig. 205 Bacfrocera correcta (Bezzi) Dacinae White & Elson-Harris 1992, fig. 179 ‘ Bacfrocera cucurbitae (Coquillett) Dacinae White & Elson-Harris 1992, ' fig. 206 Bactrocera depressa (Shiraki) Dacinae White & Elson-Harris 1992, . fig. 201 Bactrocera distincta (Malloch) Dacinae White & Elson-Harris 1992, fig. 181 Bactrocera dorsalis (Hendel) Dacinae White & Elson-Harris 1992, fig. 182 Bacfrocera facialis (Coquillett) Dacinae White & Elson-Harris 1992, fig. 183 Bactrocera frauenfeldi (Schineri) Dacinae White & Elson-Harris 1992, fig. 184 Bactrocera jarvisi (Tryon) Dacinae White & Elson-Harris 1992, fig. 175 Bactrocera kirki (Froggatt) Dacinae White & Elson-Harris 1992, fig. 185 Bactrocera minax (Enderlein) Dacinae White & Elson-Harris 1992, fig. 203 Bactrocera musae (T ryon) Dacinae White & Elson-Harris 1992, fig. 188 Bactrocera tau (Walker) Dacinae White & Elson-Harris 1992, fig. 207 Bactrocera tryoni (Froggatt) Dacinae White & Elson-Harris 1992, fig. 193 Bacfrocera fsuneonis (Miyake) Dacinae White & Elson-Harris 1992, fig. 204 Bactrocera zonata (Saunders) Dacinae White & Elson-Harris 1992, fig. 196 Callantra apicalis (Shiraki) Dacinae Shiraki 1933, pl. XIV fig. 5 Callantra ihai Shiraki Dacinae Shiraki 1968, fig. 9,10 Callantra indecora Hardy Dacinae Hardy 1974. fig. 2a Callanfra nummularia (Bezzi) Dacinae Hardy 1974, fig. 2b Callantra pedunculata (Bezzi) Dacinae Hardy 1974, fig. 3b Callantra subsessilis (Bezzi) Dacinae Hardy 1974, fig. 5b Table 10 (cont'd). 124 Species Subfamily Referenceb Ca/Ianfra vittata Hardy Dacinae Hardy 1974, fig. 6c Cerafifis punctata (Wiedemann) Dacinae White & Elson-Harris 1992, fig. 216; specimens examined Dacus abbreviatus Hardy Dacinae Hardy 1974, fig. 260 Dacus abdoangustus Drew Dacinae Hardy 1982, fig. 8a Dacus absconditus Drew & Hancock Dacinae Drew et al. 1981, fig. 3 Dacus adustus Wang & Zhao Dacinae Wang & Zhao 1989, fig. 1b Dacus aeroginosus Drew & Hancock Dacinae Drew et al. 1981, fig. 4 Dacus aethribasis Hardy Dacinae Hardy 1973, fig. 10e Dacus affinidorsallis Hardy Dacinae Hardy 1982, fig. 16a Dacus antigone Drew & Hancock Dacinae Drew et al. 1981, fig. 5 Dacus ascitus Hardy Dacinae Hardy 1983, fig. 9 Dacus auranfiacus Drew & Hancock Dacinae Drew et al. 1981, fig. 6 Dacus bangaloriensis Agarwal & Kapoor Dacinae Agarwal & Kapoor 1983, fig. 1e Dacus beckerae Hardy Dacinae Hardy 1982, fig. 17 ‘ Dacus bogorensis Hardy Dacinae Hardy 1983, fig. 10 Dacus connexus Hardy Dacinae Hardy 1982, fig. 10 Dacus costa/is (Shiraki) Dacinae Shiraki 1933, pl. ll, fig. 1 Dacus dianensis Wang & Zhao Dacinae Wang & Zhao 1989, fig. 4b Dacus disjunctus (Bezzi) Dacinae Munro 1984, fig. 41 Dacus diastafus Munro Dacinae Munro 1984, fig. 49 Dacus dispar Hardy Dacinae Hardy 1982, fig. 20a Dacus drewi Hardy Dacinae Hardy 1983, fig. 8 Dacus dubiosus Hardy Dacinae Hardy 1982, fig. 11 Dacus durbanensis Munro Dacinae Munro 1984, fig. 46 Dacus elegantulus Hardy Dacinae Hardy 1974, fig. 17c Dacus emittens Walker Dacinae Hardy 1982, fig. 12 Dacus erubescentis Drew & Hancock Dacinae Drew et al. 1982, fig. 7 Dacus flavipilosus Hardy Dacinae Hardy 1982, fig. 13 Dacus fuliginus Drew & Hancock Dacinae Drew et al. 1983, fig. 8 Dacus hyalinus (Shiraki) Dacinae Shiraki 1933, pl. I, fig. 6 Dacus involutus Hardy Dacinae Hardy 1982, fig. 23 Dacus isolatus Hardy Dacinae Hardy 1973, fig. 26a Dacus Iimbifer rufulus (Bezzi) Hardy Dacinae Hardy 1982, fig. 24 Dacus Iongisfyla Wiedemann Dacinae Hardy 1955, fig. 12 Dacus maculatus (Perkins) Dacinae Hardy 1973, fig. 270 Dacus matsumurai (Shiraki) Dacinae Shiraki 1933, pl. lll, fig. 3 Dacus melanopsis Hardy Dacinae Hardy 1982, fig. 3 Dacus montanus Hardy Dacinae Hardy 1983, fig. 7 Dacus momordicae (Bezzi) Dacinae Munro 1984, fig. 55 Dacus okunii (Shiraki) Dacinae Shiraki 1933, pl. lll, fig. 2 Dacus ortholomatus Hardy Dacinae Hardy 1982, fig. 4 Dacus perkinsi Drew & Hancock Dacinae Drew et al. 1983, fig. 11 Dacus personatus Hardy Dacinae Hardy 1983, fig. 11 Dacus petersoni Hardy Dacinae Hardy 1974, fig. 23b Dacus platamus Hardy Dacinae Hardy 1973, fig. 28c Dacus propinquus Hardy & Adachi Dacinae Hardy 1955, fig. 15 Dacus punctatifrons Karsch Dacinae Munro 1984, fig. 41 Dacus pusillus Hardy Dacinae Hardy 1983, fig. 4 Table 10 (cont'd). 125 Species Subfamily Referenceb Dacus romigae Drew & Hancock Dacinae Drew et al. 1983, fig. 12 Dacus rubiginus Wang & Zhao Dacinae Wang & Zhao 1989, fig. 1b Dacus rufofusculus Drew & Hancock Dacinae Drew et al. 1983, fig. 13 Dacus silvaticus Hardy Dacinae Hardy 1983, fig. 5 Dacus stenomus Wang & Zhao Dacinae Wang & Zhao 1989, fig. 3b Dacus sumafranus Hardy Dacinae Hardy 1983, fig. 6 Dacus tappanus (Shiraki) Dacinae Shiraki 1933, pl. ll, fig. 2 Dacus fheophrastus Hering Dacinae Munro 1984, fig. 50 Dacus transversus Hardy Dacinae Hardy 1982, fig. 6a Dacus trifasciatus Hardy Dacinae Hardy 1982, fig. 283 Dacus ubiquitus Hardy Dacinae Hardy 1974, fig. 37a Dacus vargus Hardy Dacinae Hardy 1982, fig. 15a Dacus vertebrafus Bezzi Dacinae White & Elson-Harris 1992, fig. 229 Dacus yangambinus Munro Dacinae Munro 1984, fig. 41 Dixoodacus amphoratus Munro Dacinae Munro 1984, fig. 79 Dixoodacus binotafus (Loew) Dacinae Munro 1984, fig. 81 Dixoodacus ficicola (Bezzi) Dacinae Munro 1984, fig. 86 Dixoodacus opinatus (Munro) Dacinae Munro 1984, fig. 85 Dixoodacus umbeluzinus Munro Dacinae Munro 1984, fig. 84 Ectopodacus fasciolatus (Collart) Dacinae Munro 1984, fig. 69 Ectopodacus vansomereni (Munro) Dacinae Munro 1984, fig. 71 Epochra canadensis (Loew) Trypetinae specimens examined Gymnodacus amplexus Munro Dacinae Munro 1984, fig. 35 Gymnodacus calophylli Perkins & May Dacinae Munro 1984, fig. 36 Gymnodacus kuniyoshii Shiraki Dacinae Shiraki 1968, fig. 8 Gymnodacus mesome/as (Bezzi) Dacinae Munro 1984, fig. 35 Lacfodacus adenionis Munro Dacinae Munro 1984, fig. 90 Metidacus delicatus Munro Dacinae Munro 1984, fig. 67 Mictodacus opacatus (Munro) Dacinae Munro 1984, fig. 94 Mictodacus pallidilafus (Munro) Dacinae Munro 1984, fig. 94 Myrmecodacus mirificus Munro Dacinae Munro 1984, fig. 107 Paratridacus expandens (Walker) Dacinae Shiraki 1968, fig. 12 Psi/odacus annulafus (Becker) Dacinae Munro 1984, fig. 111 Pycnodacus purpurifrons (Bezzi) Dacinae Munro 1984, fig. 119 Sfrumefa asatoi Shiraki Dacinae Shiraki 1968, fig. 9 Tomoplagia fiebrigiHendel Tephritinae Aczél 1955b, fig. 102j-k Zeugodacus ishigakiensis Shiraki Dacinae Zeugodacus scufellatus (Hendel) Dacinae Shiraki 1968, fig. 7,15 Shiraki 1968, fiL7 aA tergal pattern was judged to match the medial pattern system if one or more terga distal to syntergum 1+2 had a medial dark mark. The ceromae of dacines (see Munro 1984, p. 8) were ignored. bMunro (1984) considers dacines to constitute a separate family, the Dacidae. This view has not been adopted by other taxonomists. Table 11. Species with tergal patterns matching the sublateral pattern systema. §pecies Acanthoneura amamioshimaensis Shiraki Acidogona melanura (Loew) Acidoxantha balabacensis Hardy Acidoxanfha fofoflava Hardy Acroceratitis bimacula Hardy Bactrocera oleae (Gmelin) Campiglossa producfa (Loew) Chaetorellia acrolophi White & Marcquart Chaetore/lia conjuncta (Becker) Chaetorellia Iorioata (Rondani) Chaetorellia succinea (0. Costa) Chaetostomella cylindrica Robineau-Desvoidy Chaetostomella nigripunctata Shiraki Chaetostornella onotrophes Loew Chaetostomella undosa (Coquillett) Cryptodacus tau (Foote) Cycasia fla va Hardy Dioxyna bidentis (Robineau-Desvoidy) Dioxyna brachybasis Hardy Dioxyna sororoula (Wiedemann) Dioxyna (as Paroxyna) picciola (Bigot) Elaphromyia incompleta Shiraki Elaphromyia incompleta punctata Shriaki Elaphromyia multisetosa Shiraki Elaphromyia pterocallaeformis (Bezzi) Euaresta bella (Loew) Euaresta punctafa Shiraki Euaresta stigmafica Coquillett Eurosfa solidaginis (Fitch) Eutreta novaeboracenis (Fitch) Haywardina cuculi (Hendel) Haywardina cuculiformis (Aczél) Jamesomyia geminafa (Loew) Laksyetsa frinofata Foote Subfamily Trypetinae Tephfifinae Trypetinae Trypetinae Trypetinae Dacinae Tephflfinae Tephflfinae Tephfifinae Tephflfinae Tephfifinae Tephflfinae Tephfifinae Tephflfinae Tephfifinae Trypetinae Trypetinae Tephfifinae Tephfifinae Tephfifinae Tephfifinae Tephfifinae Tephfifinae Tephflfinae Tephfifinae Tephflfinae Tephfifinae Tephflfinae Tephflfinae Tephflfinae Trypetinae Trypetinae Tephfifinae Tephfifinae Reference Shiraki 1968, fig. 9 Benjamin 1934, fig. 24L Hardy 1974, fig. 105C Hardy 1973, fig. 101b Hardy 1973, fig. 105b White & Elson-Harris 1992, fig. 197; specimens examined Merz 1994, fig. 49; specimens examined specimens examined specimens examined specimens examined specimens examined specimens examined Shiraki 1933, pl. XI, fig. 3 specimens examined specimens examined specimens examined Hardy 1973, fig. 78b specimens examined Hardy 1988, fig. 7c Shiraki 1968, fig. 7,8 (as Ensina); specimens examined Benjamin 1934, fig. 30M; specimens examined Shiraki 1933, pl. Xl, fig. 6 Shiraki 1968, fig. 9,10 Shiraki 1933, pl. Xl, fig. 5 Hardy 1974, fig. 136b specimens examined Shiraki 1968, fig. 7,8 specimens examined specimens examined specimens examined Aczél 1951, fig. 9, 11; specimens examined Aczél 1951, fig. 21, 23; specimens examined specimens examined specimens examined Table 11 (cont'd). 127 Species Subfamily Reference Myopifes apicatus Freidberg Tephritinae specimens examined Myopifes inu/aedyssentericae Blot Tephritinae specimens examined Noeta pupil/afa (Fallen) Tephritinae specimens examined Ore/Iia falcata (Scopoli) Tephritinae specimens examined Ore/Iia occindenta/is (Snow) Tephritinae specimens examined Ore/Iia palposa (Loew) Tephritinae specimens examined Oxyna utahensis Quisenberry Tephritinae specimens examined Paracantha gentilis Hering Tephritinae specimens examined Paramyio/ia fakeuchii Shiraki Trypetinae Shiraki 1933, pl. Vlll, . fig. 3 Paraterellia immaculata Blanc Trypetinae specimens examined Paraterellia superba Foote Trypetinae specimens examined Paraterellia varipennis (Coquillett) Trypetinae specimens examined Paraterellia ypsilon Foote Trypetinae specimens examined Paroxyna absinthii (Fabricius) Tephritinae specimens examined Paroxyna albiceps (Loew) Tephritinae specimens examined; ' Jenkins 1985, fig. 124 Paroxyna clafhrata (Loew) Tephritinae specimens examined Paroxyna,difficilis Hendel Tephritinae specimens examined Paroxyna Ioewiana Hendel Tephritinae specimens examined Paroxyna matsumotoi Shiraki Tephritinae Shiraki 1968, fig. 9 Paroxyna misella (Loew) Tephritinae specimens examined Paroxyna punctata Shiraki Tephritinae Shiraki 1933, pl. XII, fig. 5 Paroxyna variabilis (Deane) Tephritinae specimens examined Phaeospilodes frifilla Hardy Trypetinae Hardy 1973, fig. 93b Rhagoletis cingulata (Loew) Trypetinae Bush 1966, fig. 50; specimens examined Rhagoletis comp/eta Cresson Trypetinae Bush 1966, fig. 63, 64; specimens examined Rhagoletis osmanthi Bush Trypetinae Bush 1966, fig. 53; specimens examined Rhagoletis zoqui Bush Trypetinae Bush 1966, fig. 61; specimens examined Rhagoletotrypeta pastranaiAczél Trypetinae Aczél 1954, fig. 7, 10 Rhagoletotrypeta xanthogastra Aczél Trypetinae Aczél 1950, fig. 3d Sophria cociinna Walker Trypetinae Hardy 1980, fig. 130 Sophria Iimbata borneensis Hering Trypetinae Hardy 1980, fig. 4b Terellia Iongicauda (Meigen) Tephritinae specimens examined Terellia Iappae (Cederhjelm) Tephritinae specimens examined Terellia ruficauda (Fabricius) Tephritinae specimens examined Terellia tussilaginis (Fabricius) Tephritinae specimens examined Terellia virens (Loew) Tephritinae specimens examined Tetramyiolia sapporensis Shiraki Trypetinae Shiraki 1933, pl. X, fig. 1 Trifaeniopferon elachispilofum Hardy Trypetinae Hardy 1973, fig. 49a Tritaeniopferon tetraspilotum Hardy Trypetinae Hardy 1973, fig. 50e Xanthomyia platyptera (Loew) Tephritinae specimens examined Table 11 (cont'd). 128 Species Subfamily Reference Xenoohaefa aurantiaca (Doane) Tephritinae specimens examined Xyphosia punctigera Coquillett Tephritinae Shiraki 1933, pl. XlV, fig. 2 Zonosemata minufa Bush Trypetinae Bush 1965, fig. 15-16 aA tergal pattern was judged to match the sublateral pattern system if one or more terga distal to syntergum 1+2 had a pair of sublateral dark marks that were closer to the midline than the lateral edges of the tergum. The ceromae of dacines (see Munro 1984, p. 8) were not counted. 129 epandrium syntergosternum 7+8 proctiger syntergosternum 6+7 surstylus blister-like structure tergum 5 sternum 5 0.25mm syntergosternum 7+8 sternum 7 sternum 6 syntergosternum 6+7 sensillum 7L ' sensillum 6R 2 0.25mm Figures 1—2. Distal abdominal structures of Rhagoletis pomonella. 1, Segments 4—8 and genitalia, ventral view. 2, Postabdominal sterna and syntergosterna, ventral view; arrow indicates point of attachment to hypandrium. 130 ejaculatory apodeme sperm sac e'aculato duct . J . ry basrphallus epandrium proctiger K prensisety/ distiphallus surstylus _ 0.25mm Figure 3. Genitalia of Rhagoletis pomonella, right lateral view. 131 ejaculatory apodeme hypandrium distiphallus phallapodeme basiphallus ’ pregonite epandrium surstylus prensiseta proctiger 0.25mm ejaculatory apodeme epandrium bacilliform sclerite hypandrium phallapodeme distiphallus surstylus basiphallus pregonite 0.25mm 5 Figures 4—5. Genitalia of Rhagoletis pomonella. 4, Ventral view. 5, Left oblique view. 132 subepandrial sclerite epandrium _-._~__ proctiger bacilliform sclerite subepandrial membrane ejaculatory duct hypoproct . _ bacilliform sclerite , 3p; prensiseta ‘5” :_~'_?_- ring-shaped ‘:. ”-‘ sclerite accessory gland surstylus hypandrium —-——-—— phallapodeme ._‘. .' . - I... . \.' .‘, l ‘ '. -' . fun , h o .2 ‘ -.“.’ v--' . ‘2 ..‘,,.m I ‘ . ‘,_';_.. \ ' I f - 3 ‘ . \ i , . . phallus I t \ n O. Q a ‘ .‘ ’ .- ‘ ’. .- 7i ____t‘-_ 1 hypandrial sac 0.1 mm 6 Figure 6. Genitalia of Rhagoletis pomonella. Sagittal section through epandrium, left lateral view. 133 subepandrial sclerite hypoproct bacilliform sclerite 0.25mm subepandrial sclerite bacilliform sclerite anterior surstylar lobe Figures 7—8. Epandrium and associated structures of Epochra canadensis. 7, Left lateral view. 8, Anterior view (proctiger omitted). Arrows indicate external sulcus (Figure 7) and internal apodeme (Figure 8). 134 subepandfial scle rite hypoproct bacilliform sclerite 0.25mm bacilliform sclerite subepandrial sclerite hypoproct anterior surstylar lobe posterior surstylar lobe 10 Figures 9—10. Epandrium and associated structures of Oedicarena Iatifrons. 9, Left lateral view. 10, Anterior view. 135 epandrium hypoproct surstylus 0.25mm 1 1 subepandrial membrane subepandnal sclerite bacilliform sclerite 12 Figures 11—12. Epandrium and associated structures. 11, Oedicarena letifrons, poste- rior view. 12, Paraterellia immaculata, left lateral view. 136 subepandfial sclerite bacilliform sclerite subepandnal sclerite anterior bridge posterior bridge bacilliform sclerite anterior surstylar lobe . . posterior surstylar lobe 15 0.25mm Figures 13—15. Epandrium and associated structures. 13, Carpomya schineri and 14, Rhagoletis cerasi, left lateral view. 15, Rhagoletis cerasi, right surstylus, medial view. 137 subepandflal sclerite bacilliform sclerite ——"\‘ \' hypoproct 0.25mm b > subepandrial sclerite / proctiger {—— bacilliform sclerite anterior surstylar lobe posterior surstylar lobe — 18 0.25mm Figures 16—18. Epandrium and associated structures of Rhagoletis berberidis. 16, Left lateral view. 17, Tip of left surstylus, lateral view. 18, Anterior view. 138 subepandnal sclerite subepandrial sclerite . bacilliform sclerite anterior surstylar lobe posterior surstylar lobe )1 0% / 2O 0.25mm Figures 19—20. Epandrium and associated structures of Rhagoletis cingulata. 19, Left lateral view. 20, Anterior view (proctiger and setae on right surstylus omitted). 139 proctiger anterior surstylar lobe / —— posterior surstylar lobe 21 0.25mm Figure 21. Epandrium and associated structures of Rhagoletis magniterebra, left lateral view. 140 subepandrial sclerite bacilliform sclerite anterior surstylar lobe . “ posterior surstylar lobe 22 0.25mm Figure 22. Epandrium and associated structures of Rhagoletis magniterebra, anterior view (proctiger omitted). 141 subepandflal sclerite bacilliform sclerite —— —— hypoproct subepandrial sclerite bacilliform sclerite . anterior surstylar lobe \ . .. posterior surstylar lobe —\\ 24 0.25mm Figures 23—24. Epandrium and associated structures of Rhagoletis psalida. 23, Left lateral view. 24, Anterior view (proctiger omitted). 142 subepandflal sclerite bacilliform sclerite , hypoproct ‘25 subepandfial sclerite ‘ "7‘?” bacilliform sclerite 1 332-7 1" .- ' ”H p ‘ / X \ -,——— anterior surstylar lobe — posterior surstylar lobe 26 0.25mm Figures 25—26. Epandrium and associated structures of Rhagoletis striatella. 25, Left lateral view. 26, Anterior view (proctiger omitted). 143 subepandfial sclerite 27 subepandrial sclerite bacilliform sclerite hypoproct posterior surstylar lobe medial surstylar lobe k} anterior surstylar lobe 0.25mm Figures 27—28. Epandrium and associated structures of Trypeta inaequalis. 27, Left lateral view. 28, Anterior view. 144 subepandnal sclerite bacilliform sclerite " subepandrial sclerite bacilliform sclerite hypoproct anterior surstylar lobe 0.25mm Figures 29—30. Epandrium and associated structures of Zonosemata electe. 29, Left lateral view. 30, Anterior view. 145 anterior surstylar lobe 1 r ‘c‘._._/\__ medial surstylar -‘F~ lobe /_" MV‘ kw era/x posterior surstylar lobe 0.1mm 31 proctiger bacilliform sclerite anterior surstylar lobe [32 Figures 31—32. 31, Tips of surstyli, Acidia cognata, ventral view. 32, Micrograph, surstyli, Rhagoletis pomonella, posterior view. 146 0.1mm 33 34 subepandrial membrane- depression 0.1mm 0.1mm epandrium basiphallus rocti er p g pregonite SUfSW'US tergum 5 Figures 33—38. 33—34, Left bacilliform sclerite, Rhagoletis pomonella. 33, Lateral view. 34, Medial view. 35—36, Prensisetae, posterior view. 35, Rhagoletis alternata. 36, Acidia cognata. 37, Left half epandrium and surstylus, Rhagoletis pomonella, medial view. 38, Micrograph, genitalia, Rhagoletis pomonella. posterior view. 147 proctiger ' «AT—subepandrial membrane 2 ., .-\ subepandrial -’ 31;;- epandrium sclerite surstylus anterior bridge \'__ posterior bridge 39 4O 0.1mm 0.1mm acrophallus accessOry gland ring-shaped sclerite . ) I I ~mWZMIU;f-\§~¥hi \lN w . . a ‘- . ..lgg :Wd E: r. {I ‘ lg M‘w’m-‘Z'L‘i-Wlwnax-a '.snm~.«..~.9; f I -.~* » ‘4‘. .' .‘ l I. .- g ...‘5 r : r basiphallus * ' . ' aedeagus I 0.25mm ' Figures 39—41, Genital structures and proctiger. 39, Bacilliform sclerites (diagram- matic), Rhagoletis pomonella, posterolateral view. 40, Proctiger (slide-mounted), Rhagoletis pomonella, ventral view. 41, Phallus, Cryptodacus tau, left lateral view. 148 basiphallic vesica lobe . apical flap appressed . flap 44 45 0.25mm 0.1 mm (inset) Figures 42—45, Distiphallus. 42—43, Paraterellia immaculata, right and left lateral views. 44—45, Oedicarena persuasa, right and left lateral-views. 149 0.1 mm (in-set) 4 7 appressed flap . -.- of‘ -- - l - .. - P Q - ..\ ‘. .' . _1.‘,‘,~ . .t.._ . .. 7f. f.f,-..I~;;-‘r.'.l~‘:‘l‘?zr '."~---.. -.\~. ...--_-‘e'¢;.-..'.-.- - ...Q .- f0 .,,.,, aedeagus .1 ' —— subapIcal lobe 48 0.25mm 0.1mm (inset) Figures 46—49, Distiphallus, right and left lateral views. 46—47, Trypeta inaequalis. 48—49, Epochra canadensis. 150 apical appressed flap \ flap subapical lobe 5 1 l4. \) acrophallus 5 O a; 1' subapical/ appressed flap fix?" {VI/Rig lobe afpiacpal 5 3 52 Jflm ‘\ flap ' ed subapical appress ((5 / bbe fl ap \(I. 5 5 5 4 0.25mm 0.1mm (inset) apical Figures 50—55, Distiphallus, right and left lateral views. 50—51, Rhagoletis adusta. 52—53, Rhagoletis cerasi. 54—55, Rhagoletis cingulata. 151 a ressed fla 1v pp p W subapical 5 7 lobe 56 1,: subapical———- QQX‘J lobe 14W “'\ appressed flap 58 aedeagus acrophallus (F ; acrophallus 6 1 apical -k.J flap 2%,. . E /I\\\ 0.25mm 6 0 ' 0.1mm (inset) Figures 56—61, Distiphallus. 56—57, Rhagoletis nova, right and left lateral views. 58—59, Rhagoletis pomonella, right and left lateral views. 60, Rhagoletis pomonella, dorsoapical view. 61, Rhagoletis magniterebra, dorsolateral view. 152 subapical ———-—;.t:_ M “4.45% “WV—A lobe =- . / ' ‘ sclerotized Icon 63 apical r\<\—’~/ Y 11;". flap ”if-1“ ‘ \ l I“\ ‘Cui ,A “‘—-.—-— subapical W lobe subapical lobe 0.25mm 0.1mm (inset) Figures 62—67, Distiphallus, right and left lateral views. 62—63, Rhagoletis psalida. 64—65, Rhagoletis striatella. 66—67, Rhagoletis suavis. 153 I. I subapical "A W lobe / . 1.2:. ”RR—t A . 51.3,}; -” apical flap €58 0.25mm j apical flap subapical lobe .\\ 15KU x399‘e\ 9914 199.90 95999 659 Figures 68—69, Distiphallus, right lateral view. 68, Chetostoma rubidium. 69, Micrograph, Rhagoletis suavis. 154 distiphallus subapical lobe apical flap basiphallus lflKU X286 @169 106.8U CEDEI'El 7O basiphallus subapical lobe 71 Figures 70—71, Micrographs, distiphallus, right lateral view. 70, Rhagoletis pomonella. 71, Rhagoletis suavis. 155 distiphallus apical flap 7_ C5039 —- parameral sheath appressed flap . . -; ,__-_— aedeagus distiphallus aedeagus —-—2 O 7 basiphallus ——— apical flap subapical lobe acrophallus C007 0 73 Figures 72—73, Phallus. 72, Rhagoletis comp/eta, apical view. 73, Phallus ground plan, right lateral VleW (cross sections: bold lines = sclerotized, plain lines = mem- branous). 156 Sc R, anterior apical band 1" Rzoa posterior apical band band dm-cu / CuA‘ b 7940qu and SC band r-m 74 Figures 74—75. 74, Generalized wihg showing venation and names of wing bands. 75, Evolution of banded wing patterns in the Trypetini. (a) Hypothetical ground plan pat- tern. (b—d) Evolution of wing pattern with proximal subcostal band prominent and band r-m complete. (e—g) Evolution of wing pattern with distal subcostal band prominent and band r-m truncated. See text. 157 . ”all? Figures 76—85. Wing patterns of trypetines. 76, Epochra canadensis. 77, Chetostoma curvinerve. 78, Euleia fratria. 79, Zonosemata e/ecta. 80, Paraterellia ypsilon. 81, Oedicarena nigra. 82. Rhagoletis pomonella. 83, Rhagoletis zoqui. 84, Rhagoletis chionanthi (normal wing shape). 85, Rhagoletis chionanthi (abnormal wing shape). Scale bars equal 1mm. 158 Figures 86—89. Wing patterns of Rhagoletis. 86, Rhagoletis fausta (normal wing shape). 87, Rhagoletis fausta (abnormal wing shape). 88, Rhagoletis jug/andis (normal wing shape). 89, Rhagoletis jug/andis (abnormal wing shape). Scale bars equal 1mm. 159 band lost posterior apical band lost secondary division and reduction of anterior apical band anterior apical band lost bands h and so coalesce 90 Figure 90. Transformation series for wing patterns in Rhagoletis. (a) Rhagoletis blanchardi, (b) Rhagoletis striatella, (c) Rhagoletis cerasi, (d) Rhagoletis comp/eta, (e) Rhagoletis indifferens, (f) Rhagoletis cingulata. (g) Rhagoletis tabellaria, (h) Rhagoletis zephyria. Circles on vein R4+5 indicate position of campaniform sensilla. Scale bars equal 1mm. Drawings of wings were made using a drawing tube attached to a stereo microscope. 14.9. 160 0 R. indifferens O R. cingulata O Oom®®cooooo 0'0 000:) @00 "c C CD .0 _ complete - <13 .2 o. 03 q— o c .9 9; broken - “o c o o 0.3 I I I I I I I I 0.4 0.5 0.6 0.7 0.8 0.9 1 1.1 distance A / distance B Figure 91 . Relationship between condition of apical band and ratio of distance between the two distal sensilla on vein R4+5 (distance A) to the distance between the distal most sensillum and apex of vein R4+5 (distance B) in R. cingulataand R. indifferens. 161 1 5 H U H 4 4 l3 Figures 92—93. Scanning electron micrographs of setae of Rhagoletis species. 92, Right orbital setae of R. pomonella showing longitudinal grooves. 93, Right upper frontal seta of R. comp/eta showing oblique striations lying in longitudinal grooves. 162 Figures 94—95. Scanning electron micrographs of scutal microtrichia of Rhagoletis pomonella. 94, Microtrichia from lateral microtrichiose stripe. 95, Microtrichia from between sublateral and lateral microtrichiose stripes. ”2:324 R, juglandis (8) 3555535 Rh. pastranai (2) a R. reducta (3) . 5:321: H. cuculifonnis ( 1) 23955223 T, toru‘le ( 1) 15355571 fractura (1) 0' l°8£ 553353; 0 ' |-' l L ' 39' :;s§e§s§ 0 . L‘SL' 313535? 0' 1'91? . 0' 1‘09' 5.51.492: 2. 3 5 ‘ ' Ll.’ £2332 . 89"99 '55- 0 ' L‘El. 68"8L' eggs; 0' l'CL' 3:22;; 2 6 '- 8L' 13355:; 68"0L' 333;}; 3»? i g . turanicum (7) . 'florida' (8) chionanthi ( 8) acuticarnis (3) ferruginea (5) . caucasica (2) perfecta (4) wiedemenni (5) pomonella (8) intermedie (4) . . inaequalis (6) 68.11.. .. Iongipennis (4) 06:81.. 33255 25555525 Rh. annulata (4) 06 ' l»!— 111“ R. suavis (8) 16' 1'9 ’95 5:32;: FI. striatella (8) . meigenii (3) R. blanchardi (6) - . kurentsovi (5) ' '33? Ch. curvinerve (4) . R. tomatis (8) 68:19 17.31-32.55?! =9?" R. chopersella (8) 9,119- ;;‘::e . R. a. orientale (2) 09:18.8. altemata (8) 411-69. 2333235.?" 9* E. canadensis (8) 94 :39 . “ ' Cr. tau (6) 99:99. Y? 0. nigra (4) 91:39. _3::.': cl 0, Iatifrons (4) V9.’99. " ::-. My. lucida (2) 89:99 A, cognate (5) OL.’09 , a P, ypsilon (4) ’9.‘79. '11 My, Iimata (4) ' ”1'39. R. basiola (8) Eli-1.17 o, tetanops (3) 09:99 0. persuasa (2) 99:97 P. superba (4) 89 -OS P. immaculata (5) . 99 0. beameli (1) 09-09 z. vittigera (4) 19"9" - z. electe (4) 09"” . ' P. varipennis (3) 817:9? ; Z_ vidrapennis (2) I I '87 '3]? ‘ z. scutellum (3) «a " "- "- ° N N "' 0 mean spal s / dc s £95302: 163 C. schineri (4) R. rhytida (2) R. flavicinta (6) --’ ’41” R. metal/ice ( 1) 9 unfit/’fe‘ficm’fi R. obsoleta ( 1) . mango/ice (2) R. ribicola (8) . nova (8) . tabellarla (8) .. - ... . nr. tabelleria (8) 15435162311549 n, cerasi (8) O .‘::'.'.‘>:1,.5" ’4’" 1113‘ .- .;;. My” 9 3' 911- o’fiflfififi‘Wi 9 111:9 OZZI 1'.” ‘::.... ”n“... ' 55"”9n9eewwsés R SL.L'LI1L§:::§:§E 1. L9'l‘3 88'l'53'l5-t- Wifi C, vesuviana (2) :1? >' 1:1: 46'. ,3} R S"'3:l0l SL'LL' <19? . fausta (8) . berberis (8) . batava (8) R. ramosae (2) 93 3. 93' 1 {£35 ‘1 EV'l'Sl'l :ii': . L l '1 , “69$ -3,R scutal/eta (1) 634-0 2.9%.» R. macquertii (5) 88 l‘0 5;??? ‘91,)? magniterebre (5) 98'1'0' .. 111.59 fratria (5) 2° I-0' I» 3' L'88' 5:113:13 ZZ'L'O' 99'1'16 3333:; u'.‘ 'lo'u'. ....':_._ ------------- V P P I 0 [x {3232}; 2:223; n, complem (8) 33:2: .53.: R. adusta (2) :i:€:i;3.- n, zoqui (8) 513533? R. zernyi (2) ';.:r:'- H cuculi (4) 01.68 Ch. californiann (4) a. ". ° ,_ o I w ‘- N N mean spal s / do 3 Figure 96. Mean ratio of distance, measured from transverse suture, of the supra-alar seta (spal s) to the dorsocentral seta (dc 3). Number in parentheses after species is sample size; range of the ratio is given above bars. See text. 1 6 4 SS. “87 "'##3‘9':1:553”??? R, berberis (8) 56"98. :CITSEEE‘I; ........................................................... 5,333 Cf. tau (6) ' . adusta (2) t8'-8/.' 351 H. CHOU” (4) . incompleta (4) . 55133 My. nigricornis (1) . suavis (8) 6’- :89 if??? *1: Eu heraclei (4) .pomonella (a) 3’- '°’- s Intermedia (4) . osmanthi (a) _69. =11 T. fracture (1) . lycoperselle (8) 910-79.: 353 S. Iongipennis (4) n. juglandis (8) ”- :99 , 311* Eu. uncinata (4) 15?: n. comp/eta (a) 91.-19.»... 1* 8- perfecta (4) 5555351. chionanthi (a) ”:39 , 11??? 15*? Eu. fratria (5) 551510. tetanops (3) L9 :19 ‘5‘: 535 114% Iimata (4) 1:3 69 '99 352‘ -- 5355 My. lucida (2) 29..09- ............ Ch. californicum (4) E22: E. canadensis (8) .zemyi (2) . Z ephyna (8) $9 ’69 :;§Z:1:§:§ij."‘_. '3, 09~695,g7 a) c: c (U L— 6 .E a) .9 o. E (U rn .9 Q E to 1:: fl) .9 o a) o. In . reducta (3) W19, P. ypsilon (4) g . nr. tabelleria (a) ., P- ”mama“ 15’ a . mango/ice (2) Ch. rubidium (3) 8 . flavigenualis (5) 09'-ZS' 2' scutal/are (3) 3 . comivom (a) 99., ”.75: if; P. varipennis (a) *‘5 . acuticarnis (3) 09.,99. R‘ 20‘7“] (8) 93 . 'flon'da' (8) 99f A. cognate (5) g . persuasa (3) VS' H. cuculiformis (1) . .E . wiedemenni (5) ,9. ............... if T‘ “”1”" (.1) 8 . striatella (a) £9509. if R- “be“ (1) D caucasica (2) 1°" 1": P1 9UP9’P3 (4) E ‘ . 69"5‘” 393: T inaequalis (6) 3 . almatensrs (2) 79"89' ; ,2, ’ Z . Iatifrons (4) tsulg' 112' ”M’s?" (2)2 2' , : 2+: . vesuvrana ' pardalina (8) 99"”. : 5323'? C. schineri (4)( ) .E - gadgfz) EST”: ’15 R. tomatis (8) g _, ' , , a 99 °9 552:? n. indfferens (a) a: ..:-: . ”19199", ( ) 1.9“99' $5.} R. OOMi (8) s ff R- ”8“” (a) 231251 blanchardi (6) 7“ *1: Ch. CUMMIVG (4) 39"67. 3.13:9: 1E1: . ' ' E ' 5395 Rh. rohweri (4) IS' ‘11-: 1* Z vmrgera (4) "' T" 33: mendax (8) . 1‘11 Rh. pastranai (2) c 9 . . 35 ‘09 ’3??? 3555 Rh. annulata (4) '5 o . magnrterebra (6) 79"97' 33;; 55;; R. ribicola (8) 5 0 . fausta (8) ZS'-8V .1351; 31533 R. flavicinta (6) 8 w . CODVOFSR (7) 89"87. 513533; iii; R. ferruginea (6) 8 as .beamen (1) -- "' a: , ,, . . R. emiliee (1) o n . persimilis (8) 99 '97 i952? 5:35? B. cingulete (8) c m . a. Grimm’s (2) 99.‘L17' 511:1 3:}; R. boycei (8) «g 8 . tabelleria (8) 89"97'51235 {53.335 R. berberidis (3) g a . altemata (8) l9"6V as: (555': z, vidrapennis (2) 0’ c . kurentsovi (5) 39"57° :55; 92:? z, electe (4) 5 .g . electromarpha (8) . 09. 533;; ;23;;' R. turanicum (1) g :1 . b88l0l8 (8) 39 '31? jsgég': 3323‘; R. psalida (8) “5 ‘1: R jamaicensis (7) 09: :f-zis iéiii? n, penela (1) 3 to; O. nigra (4) . 09_ 5.55? 555? R. metallica (1) o g l l l R macquam' (5) 95 '4’ ::':,§12Ef::; .......... 2212:1131 n. juniperina (3) 5 ,5 v- “) ID. ID 0 7- ID ID IO 0 .9 ..- " o “l ". o' e! “- ° 0 o o 0 mean distance from bm-cu to r-m / distance from bm-cu to dm-cu ‘ 165 N . ,1 .-‘. 6 P" ;-1 ‘21? .1." .;- <1 :55 1,31: z z-o , O . oooooQR I I I to 0 I!) F F OJOOOOOOOOOOOOOOOOOOOOOO mean number of setae on R4+5 R. adusta (2) R. ferruginea (6) R. nova (8) R. blanchardi (6) R. jamaicensis (7) . conversa ( 7) . striatella (8) . lycoperselle (8) . macquam'i (5) . psalida (8) . basiola (8) . rhytida (2) . caucasica (2) . magniterebre (6) . zoqui (8) . meigenii (8) . metallica . emiliee ( 1) complete (8) boycei (8) . altemata (8) . juglandis (8) ramosae (2) penela (1) acuticomis (3) cerasi (8) berberidis (8) kurentsovi (5) flavigenuaIis (5) ribicola (8) reducta (3) pomonella (8) juniperina (8) indifferens (8) zernyi (2) zephyria (8) turanicum ( 1) tabelleria (8) suavis (8) persimilis (8) . osmanthi (8) . nr. tabelleria (8) . mongoiica (2) . mendax (8) . fiavicinta (6) . fausta (8) . electromarpha (8) . ebbettsi ( 1) . cornivora (8) . cingulete (8) . chionanthi (8) . berberis (8) . batava (8) . almatensis (2) . a. orientale (2) . 'fion’da' (8) Figure 98. Mean number of setae on vein R4+5 dorsally beyond branching 01 V91" Rs in Rhagoletis. Number in mmmmmmmmmmmmmummpmmmmmmmmpmpmmpmmmmmmmmmmmmmmmmmmmm parentheses after species is sample size; range in number of setae is given above bars. L'O R if??? p ,4. R 32:7:- R 5212333 R 65555} F? 1:52:53 7?? R ,3 R 7 T R G R V R j R R In M R R. R R. R R. R R R R. R R R. R. R. R. R. R. R. R. R. R. R. R. R. R. R. P. C. C. C. l'O I'O l'O O OOOOOOOOOOOOOOOOOOOOOOOOO . meigenii (8) . varipennis (3) . metallica ( 1) . emiliee ( 1) . complete (8) . boycei (8) . altemata (8) . juglandis (8) . ramosae (2) . tortile ( 1) . penela ( 1) . wiedemenni (5) . acuticarnis (3) . cerasi (8) . berberidis (8) . kurentsovi (5) . flavigenuaIis (5) . pardalina (8) . ribicola (8) reducta (3) . pomonella (8) juniperina (8) . indifferens (8) zemyi (2) . zephyria (8) . turanicum ( 1) . tabelleria (8) suavis (8) . persimilis (8) . osmanthi (8) nr. tabelleria (8) mongolica (2) mendax (8) flavicinta (6) fausta (8) electromarpha (8) ebbettsi ( 1) cornivora (8) cingulete (8) chionanthi (8) berberis (8) batava (8) almatensis (2) a. orientale (2) 'flon’da' (8) ypsilon (4) veswiana (2) schineri (4) incompleta (4) I I I T 0 l0 0 In N v— ‘- mean number of setae on R4+5 O 166 0 a $33,355,555, ......................................................... My nigricornis (7) 83' l l Rh pastranai (2) 61‘“ ..'.'..1221.I.Z222-3925 Ch. californicum (4) 5 1‘" 1‘ W 55555 . vidrapennis (2) 8 l-0 L 5:35:52? Eu heracleu (4) L l' L L ..w‘ Eu. frame (5) ,3; 23259319999351 E. canadensis (8) 92611193999" A. cognéItal (5) >35: .W/fiffir‘efifis z. electa (4) ' z scutal/ate (3) ’23.? ”,0, tau ( 6) “Ky-5.x, R jamaicensis (7) . interrnedia (4) . Iongipennis (4) . perfecta (4) . conversa ( 7) . cuculi (4) . striatella (8) . lycoperselle (8) . superba (4) . toma tis (8) . basiola (8) . beamen' ( 1) » . cuculiformis (7) ”2%5: T, inaequalis (6) 2:120. nigra (4) 2,2.;:%_1§,;.;_};,i:§ 0, tetanops (3) "" . immaculata (5) . persuasa (3) . rhytida (2) . caucasica (2) 0. Iatifrons (4) , magniterebre (6) . zoqui (3) 15- 10... l o 0] mean number of setae on R4+5 Figure 99. Mean number of setae on vein R4 +5 dorsally beyond branching of vein Rs. Numbers in parentheses after range for number of setae is given above bars. species is sample size 167 MEDIAL PATTERN SUBLATERAL PATTERN SYSTEM SYSTEM L J L J I I I f I ' Figure 100. Transformation series for medial and sublateral pattern systems of tergal coloration. See text. . percent of subfamily 168 1:1 Sublateral Symmetry System I Medial Symmetry System 100- .99 96.3- 93 75- 50- 25~ _ 0- 0'9 A #— Dacinae (117) Trypetinae (27) Tephritinae (51) Fig. 101. Percent of species by symmetry system within subfamilies. Number in parentheses after subfamily is sample size. See text. 169 Figures 102—103. Scanning electron micrographs of denticles on eversible ovipositor sheath of Rhagoletis species. 102, R. cornivora. 103, R. pomonella. Carpomyina “ unplaced genera Trypetina ewe/e 'z simian/none 'H ,I/nona 'H [JquOJ 'qy ere/nuue my net '19 ,ueumos '9 age/dwoou! '9 ,Iuefijeiu 'u _mosruejnx 'H [euansed 'qy euyepjed ,quewepepn ,IA'UJez eye/[eqaz smens 9119121113 a/oayqu epyesd elleuowod smugsmd eel/ofiuow equJeI/ufiew [zjanboeui alouioi/Iey aculfinue) BJSHB} BIO/[IUJOO BSJOAUOO aze/dwoo era/nfiuia ,IseJeo sueqleq 5!P!J9qJ9q newer/a slums/Inca 110/!de quedns ale/noewwj 'tL'l'n.‘ 05050505051:mmmmmtfaiufaiuim'cfinftftrct'uici: e/ueuy/uno 'qo Lunolwomao ‘qo apronl 'Kw 319w” Kw siuuedlb‘uol '9 ejpeuueju! 's ejeuiaun 'n3 811)?!) 'n_:] syenbeeui 1 creation 17 s/suepeueo '3 39 170 = 0.417, and RI = 0.748 (uninformative 135, CI Bars represent synapomorphies; numbers refer to characters in Table 7. Subtribes are indicated along the top of the tree. Figure 104. Strict consensus tree of 18,691 cladograms of length characters ignored). Carpomylna unplaced genera Trypetlm 3190mm '2 aluuedupln 'z DWI/91003 'Z I, . para 2 syuuou/nana 'H none 'H W 'H Iii/{WIN9d 'H uou 14 ”10500“! 'H xepueul 'u uqaaeuuflaw y [unnbomu 'u WW0“! ”H .vpyou. ' 811'"!!me ‘u mammal; ‘5 man; 'u 3399!!“ 's slwedlfiuw 's Waylon} srsuapeuea '3 171 0.848. Bars represent 0.430, and RI synapomorphies; numbers refer to characters. Subtribes are indicated along the top of the tree. 135, CI Figure 105. Strict consensus tree of 1,000 cladograms of length Carpomyina Trypetina unplaced genera F ezoele 'z swung/none ‘H .Ilnono 'H [JOM qo: 'qu ale/nuue 143 8919/! uoo ' guenba ew eeuyfinue} ' [Jeurqos eJe/dwoou! enemas ' [A OSJUQJM ezeweye swans age/dame ' ere/nfiugo ' euyepjed w \ 40 IDIOZ'I'EQUO'QUI'O: e/[euowod “u exonguxoa y leueased 'qy Mmez '5 9."l 9119qu 9709M.“ swarmed ' eoyofiuow ruetilew equJeyufiew BJOUpM e]; egsne; _IseJeo sueqjeq S!P119419<7 _ slumognoe * exeupun 'n3 0 3mm; 'n3 ewufioo “v sluuedlbuol 's egpeuuezu! '9 e/ueupuno 119 c» wnarummea 'qo anion: WW ewwy 'Kw syenbeeu! 1 N uousdlf 'd 0 eqredns ‘d axe/no eww! 'd c, co esenSJed '0 “‘ 0 amp '0 it 0: niaiaz'tt‘m'az'afaicfi 10 69 52’ L summer '0 srsuepeuea '3 [uuewepem '9 / g 172 65 48 24 56 55 0.671, and RI = 0.900 (reweighted Figure 106. Strict consensus tree of 12,061 cladograms of length = 37,433, CI data). Bars represent synapomorphies; numbers refer to characters in Table 7. Subtribes are indicated along the top of the tree. Carpomyina unplaced genera Trypetina r [encased 'qu sum: 1; W 11 Up. #1 'u any/99d U BAOU ‘U ewe/moo y manna; '3 2:3an '3 platinum 'H 8 ea ' [an '95 E swears”! 'u murmur?! '3 3mm 'H spur/fin! 'u mboz '3 989041181 g mordww : U muewso y “WHIP"! “U memb‘up ' MWOflP ' udeZ' "guMoqud S xepuew y .PPMOIL 'H mules '5 WM '8 MW! '8 ‘IIWUOQMPII 'u mean ' .1190!!!” '0 weldwoour '9 equepmd w !W3WOP9.W '9 9.0311948! '8 WWWIO 'u egsne; y 9.03””?! W 'H summed 'u [sens y sysuezewle y SIPIJOQJW 'H _rueflrew 'u Mosauemx '5 "loan“ 'H coyofiuow 'y uqemwflew 'u BJOUPM PI! ’8 caisson” '3 8mm ‘8 HOW 'H lawsuit '8 #0100909? ‘U alumna" 'UU WWI "(H emnuue ya me '10 ”none ‘H PJ'IIOJDOS 'Z 339919 7 WWOUPPM '2 medium '2 qummnona 'H * uogsdA 'd sruuedyen 'd emedns 'd emmewwg 'd sdouwe: '0 esensred '0 316m '0 suolmel '0 emuwno 149 «WWW/99 'uo app"! m mm.” “M opened '3 qwedgfiuol 's WPOWJOJU! 'S ewqraun ’n3 WWW '03 BMW 71.? sgenbeeu; 1 p- eieufioa 'v sgsuepeuea '3 4O 26 48 173 9\ 25 4 16 82 a ‘0 [s 55 10 39 H 0.789. Bars represent 0.281, and RI numbers refer to characters in Table 7. Subtribes are indicated along the top of the tree. Figure 107. Strict consensus tree of 13,100 cladograms of length = 306, CI synapomorphies number of polymorphic species 25" 204 15' 10" 174 3364731421304546 811313536382768743423 619431 28 718 5 4 character number Figure 108. Number of polymorphic species by character. LIST OF REFERENCES LIST OF REFERENCES Aczél, M. L. 1950. A revision of the genus 'Xanthaciura" Hendel (Trypetidae. Dipt.) based on Argentine species. Acta Zool. Lilloana 8 (1949): 111—146. Aczél, M. L. 1950. Generos y especies de la tribu “Trypetini". l Dos generos y tres especies neuevos de la Argentina (Tephritidae, Diptera). Acta Zool. Lilloana 9: 307—323. Aczél, M. L. 1951. Generos y especies neotropicales de la tribus 'Trypetini'. ll Dos generos y una especies neuevos. Acta Zool. Lilloana 12: 253—278. Aczél, M. L. 1952. Further revision of the genus Xenthaciura Hendel (Trypetidae, Dipt.). Acta Zool. Lilloana 10: 245—280. Aczél, M. L. 1954. Generos y especies de la tribus Trypetini 4: El genero Rhagoletotrypeta y nuevas especies de Tomoplagia y de Zonosemata (Diptera, Tephritidae). Dusenia 5(3—4): 137—164. Aczél, M. L. 1955a. The comparative morphology of the Tomoplagia Coquillett species ‘ (Diptera, Trypetidae). Dusenia Vl(5): 139—170. Aczél, M. L. 1955b. Fruit flies of the genus Tomoplagia Coquillett (Diptera, Tephritidae). Proc. U. 8. Nat. Mus. 104(3343): 321—411. Agarwal, M. L. and V. C. Kapoor. 1983. Two new species of Dacus Fabricius (Diptera: Tephritidae) from India. J. Entomol. Res. 7(2): 169—172. Bainbridge, S. P., and M. Bownes. 1981. Staging the metamorphosis of Drosophila melanogester. J. Embryol. eXp. Morph. 66: 57—80. Barber, E. W. 1994. Women's work: The first 20,000 years. W. W. Norton and Co., New York. 334 pp. Bard, J. B. L. 19977. A unity underlying the different zebra striping patterns. J. Zool., Land. 183: 527—539. Benjamin, F. H. 1934. Descriptions of some native trypetid flies with notes on their habits. USDA Tech. Bull. 401, 96 pp., Washington, D. C. Berlocher, S. H. 1981. A comparison of molecular and morphological data and phonetic and cladistic methods, in the estimation of phylogeny in Rhagoletis (Diptera: Tephritidae), pp. 1—31. In M. W. Stock [ed.], Application of genetics and cytology in insect systematics and evolution. Forest, Wildl. and Range Exp. Sta., Univ. Idaho, Moscow. 152 pp. 175 176 Berlocher, S. H. 1984. A new North American species of Rhagoletis (Diptera: Tephritidae), with records of host plants of Cornus-infesting Rhagoletis. J. Kans. Entomol. Soc. 57: 237—242. Berlocher, S. H., and G. L. Bush. 1982. An electrophoretic analysis of Rhagoletis (Diptera: Tephritidae) phylogeny. Syst. Zool. 31: 136—155. Berlocher, S. H., B. A. McPheron, F. L. Feder, and G. L. Bush. 1993. Genetic differentiation at allozyme loci in the Rhagoletis pomonella (Diptera: Tephritidae) species complex. Ann. Entomol. Soc. Am. 86: 716—727. Berube, D. E. 1978. Larval descriptions and biology of Tephritis dilacerata (Dip.: Tephritidae), a candidate for the biocontrol of Sonchus arvensis in Canada. Entomophaga 23: 69—82. Bierbaum, T. J. and G. L. Bush. 1990. Genetic differentiation in the viability of sibling species of Rhagoletis fruit flies on host plants, and the influence of reduced hybrid viability on reproductive isolation. Entomol. exp. appl. 55:105—118. Boller, E. F. and R. J. Prokopy. 1976. Bionomics and management of Rhagoletis. Ann. Rev. Entomol. 21:223—246. Bryant, H..N. 1989. An evaluation of cladistic and character analyses as hypothetico- deductive procedures, and the consequences for character weighting. Syst. Zool. 38: 21 4—227. Bush, G. L. 1965. The genus Zonosemata, with notes on the cytology of two species (Diptera—Tephritidae). Psyche 72: 307—323. Bush, G. L. 1966. The taxonomy, cytology, and evolution of the genus Rhagoletis in North America (Diptera, Tephritidae). Bull. Mus. Comparative Zool. 134(11): 431—562. ' Bush, G. L. 1992. Host race formation and sympatric speciation in Rhagoletis fruit flies (Diptera: Tephritidae). Psyche 99:335—358. Campuzano, S., and J. Modolell. 1992. Patterning of the Drosophila nervous system: the achaete-scute gene complex. Trends Genet. 8: 202—208. Carpenter, J. M. 1988. Choosing among multiple equally parsimonious cladograms. Cladistics 4: 291—296. Cole, F. R. 1969. The flies of western North America. Univ. Calif. Press, Berkeley and Los Angeles. xi + 696 pp. Condon, M. A., and A. L. Norrbom. 1994. Three sympatric species of Blepharoneura (Diptera: Tephritidae) on a single species of host (Gurania spinulosa, Cucurbitaceae): new species and new taxonomic methods. Syst. Entomol. 19: 279— 304 Cumming, J. M. , B. J. Sinclair, and D. M. Wood. 1995. Homology and phylogenetic implications of male genitalia in Diptera—Eremoneura. Entomol. Scand. 26: 120— 1 51 . 177 Davis, J. l., and K. C. Nixon. 1992. Populations, genetic variation, and the delimitation of phylogenetic species. Syst. Biol. 41:421—435. Doyle, J. J. 1992. Gene trees and species trees: molecular systematics as one- character taxonomy. Syst. Bot. 17:144-163. Drew, R. A. I. 1969. Morphology of the reproductive system of Strumeta tryoni (Froggatt) (Diptera: Tephritidae) with a method of distinguishing sexually mature adult males. J. Aust. Entomol. Soc. 8: 21—32. Drew, R. A. I. 1972. The generic and subgeneric classification of Dacini (Diptera: Tephritidae) from the South Pacific area. J. Aust. Entomol. Soc. 11: 1—22. Drew, R. A. l., D. L.- Hancock, and M. C. Romig. 1981. Australian Dacinae (Diptera: Tephritidae): New species from Cape York Peninsula, a discussion of species complexes and Key to species. Aust. J. Zool. 29: 49—91. Eberhard, W. G. 1990. Animal genitalia and female choice. Am. Scientist 78:134— 141 . _ *1. Evans, H. E. .1977. Extrinsic versus intrinsic factors in the evolution of insect sociality. Bio. Sci. 27:613—617. Farris, J. S. 1969. A successive approximations approach to character weighting. Syst. Zool. 18: 374—385. Feder, J. L., C. A. Chilcote and G. L. Bush. 1988. Genetic differentiation between sympatric host races of the apple maggot fly Rhagoletis pomonella. Nature 336(6194): 61—64. Foote, R. H. 1959. Notes on the genus Euleia Walker in North America (Diptera: Tephritidae). J. Kansas Entomol. Soc. 32(4): 145—150. Foote, R. H. 1980. Fruit fly genera south of the United States (Diptera: Tephritidae). U.S. Dept. Agri. Bull. 1600 ‘ Foote, R. H. 1981. The genus Rhagoletis Loew south of the United States (Diptera: Tephritidae). US. Dept. Agri. Bull. 1607. Foote, R. H. 1984. Family Tephritidae (T rypetidae), pp. 66—188. In A. Seas and L. Papp [eds.], Catalog of Palearctic Diptera, vol. 9 (Micropezidae—Agromyzidae). Akademiai Kaido, Budapest, Hungaray. 460 pp. Foote, R. H., and G. C. Steyskal. 1987. Tephritidae. Pages 817—831 in Manual of Nearctic Diptera, vol. 2 (J. F. McAlpine, B. V. Peterson, G. E. Shewell, H. J. Teskey, J. R. Vockeroth, and D. M. Wood, coords). Res. Br., Agric. Can., Monogr. 28, Ottawa. Foote, R. H., F. L. Blanc, and A. L. Norrbom. 1993. Handbook of the fruit flies (Diptera: Tephritidae) of America north of Mexico. Comstock Publishing Associates, Ithaca, N.Y. xii + 571 pp. 178 Forey, P. J., C. J. Humphries, l. J. Kitching, R. W. Scotland, 0. J. Siebert, and D. M. Williams. 1992. Cladistics. A practical course in systematics. Oxford University Press, New York. x + 191 pp. Freidberg, A. 1980. A revision of the genus Goniure/lia Hendel (Diptera: Tephritidae). J. Ent. Soc. Sth. Afr. 43: 257—274. Freidberg, A. 1994. A second species of Notommoides Hancock 1986, with a re- description of the type species (Diptera Tephritidae Typetinae). Trop. Zool. 7: 333—341 . Freidberg, A., and W. N. Mathis. 1986. Studies of Terelliinae (Diptera: Tephritidae): A revision of the Genus Neaspi/ota Osten Sacken. Smithsonian Contrib. Zool. no. 439. 75 pp. Frey, J. E. and G. L. Bush. 1990. Rhagoletis sibling species and host races differ in host odor sensitivity. Entomol. exp. appl. 57:123—131. Frias, D., M. lbarra, and A. M. Llanca. 1987. Un nuevo disefio alar en Rhagoletis. conversa (Brethes) (Diptera: Tephritidae). Rev. Chilena Ent. 15: 21—26. Garcia-BellidosA" P. A. Lawrence, and G. Morata. 1979. Compartments in animal development. Sci. Amer. 241: 102—110. Ghiradella, H. 1984. Structure of iridescent lepidopteran scales: variations on several themes. Ann. Entomol. Soc. Am. 77:637-645. Griffiths, G. C. D. 1972. The phylogenetic classification of Diptera Cyclorrhapha with special reference to the structure of the male postabdomen. Series Entomologica, vol. 8 (E. Schimitschek, ed.). Dr. W. Junk N. V., The Hague. 340 pp. Han, H.-Y. 1992. Classification of the tribe Trypetini (Diptera: Trypetinae: Trypetini). Ph. D. dissertation, Pennsylvania State University. Han, H.-Y., and B. A. McPheron. 1994. Phylogenetic study of selected tephritid flies (lnsecta: Diptera: Tephritidae) using partial sequences of the nuclear 183 ribosomal DNA. Biochem. Syst. Ecol. 22: 447—457. Han, H.-Y., and B. A. McPheron. submitted. Molecular phylogenetic study of Tephritidae (lnsecta: Diptera) using partial sequences of the mitochondrial 16s ribosomal DNA. Molec. Phylo. Evol. Han, H.-Y., X.-J. Wang, and K. C. Kim. 1993.1 Revision of Cornutrypeta Han & Wang, a new tephritid genus proposed for Oriental and Palearctic species (Diptera: Tephritidae). Ent. scand. 24: 167—184. Han, H.-Y., X.-J. Wang, and K. C. Kim. 1994a. Paratrypeta Han & Wang, a new genus of Tephritidae (Diptera) from China. Orient. Insects 28: 49—56. Han, H.-Y., X.-J. Wang, and K. C. Kim. 1994b. Taxonomic review of Pseudina Malloch (Diptera: Tephritidae) with descriptions of two new species from China. Orient. Insects 28: 103—123. 179 Hancock, D. L. 1986. Classification of the Trypetinae (Diptera: Tephritidae), with a discussion of the Afrotropical fauna. J. Entomol. Soc. S. Afr. 49:275—305. Hardy, D. E. 1955. A reclassification of the Dacini (Tephritidae—Diptera). Ann. Entomol. Soc. Am. 48(6): 425—437. Hardy, D. E. 1973. The fruit flies (Tephritidae—Diptera) of Thailand and bordering countries. Pacific Insects Monograph 31: 1—353. Hardy, D. E. 1974. The fruit flies of the Philippines (Diptera: Tephritidae). Pacific Insects Monograph 32: 1—266. Hardy, D. E. 1977. Family Tephritidae (Trypetidae, Trupaneidae), pp. 44—134. In M. D. Delfinado and D. E. Hardy [eds.], A catalog of the Diptera of the Oriental Region. vol. 3, Suborder Cyclorrhapha (excluding Division Aschiza). Univ. Hawaii Press, Honolulu. x + 854 pp. Hardy, D. E. 1980. The Sophira group of fruit fly genera (Diptera: Tephritidae: Acanthonevrini) Pacific Insects 22(1—2): 123—161. Hardy, D. E. 1982. The Dacini of Sulawesi (Diptera: Tephritidae). Treubia 28(5): 173—241 . Hardy, D. E. 1983. The fruit flies of the genus Dacus Fabricius of Java, Sumatra and Lombok Indonesia (Diptera: Tephritidae). Treubia 29(1): 1—45. Hardy, D. E. 1986. Fruit flies of the subtribe Acanthonevrina of Indonesia, New Guinea, and the Bismark and Solomon Islands. Pacific Insects Monograph 42: 1—191. Hardy, D. E. 1988. The Tephritinae of Indonesia, New Guinea, the Bismark and Solomon Islands (Diptera: Tephritidae). Bishop Mus. Bull. Entomol. I. Bishop Museum Press, Honolulu. vii + 92 pp. Hernandez-Ortiz, V. 1985. Descripcion de una nueva especie Mexicana del genero Rhagoletis Loew (Diptera: Tephritidae). Folia Entomol. Mex. 80:73—79. Hernandez-Ortiz, V. 1993. A new species of Aciurina Curran (Diptera: Tephritidae) from Oaxaca, Mexico. Folia Entomol. Mex. 87: 49—54. Hernandez-Ortiz, V. 1993. Description of a new Rhagoletis species from tropical Mexico (Diptera: Tephritidae). Proc. Entomol. Soc. Wash. 95:418—424. Herring, E. M. 1947. Bestimmungstabelle der Unterfamilien und Tribus der Trypetidae. Siruna Seva 6:12—16. Jenkins, J. 1985. A revision of the North American Tephritis species attacking Baccharis with an evaluation of morphological characters. MS thesis. Washington State University, Pullman. 132 pp. Jenkins, J. 1990. Mating Behavior of Aciurina mexicana (Aczél) (Diptera: Tephritidae). Proc. Entomol. Soc. Wash. 92: 66—75. 180 Jenkins, J. and W. J. Turner. 1989. Revision of the Baccaris-infesting (Asteraceae) fruit flies of the Genus Tephritis (Diptera: Tephritidae) in North America. Ann. Entomol. Soc. Am. 82(6): 674—685. Johnson, P. A., F. C. Hoppensteadt, J. J. Smith, and G. L. Bush. 1996. Conditions for sympatric speciation: a diploid model incorporating habitat fidelity and non-habitat assortative mating. Evol. Ecol. 10:187—205. Johnson, S. A. and M. J. Milner. 1987. The final stages of wing development in Drosophila melanogester. Tissue & Cell 19(4): 505—513. Kamali, K., and J. T. Schulz. 1974. Biology and Ecology of Gymnocarena diffuse (Diptera: Tephritidae) on sunflower in North Dakota. Ann. Entomol. Soc. Am. 67: 695—699. Kluge, A. G. and A. J. Wolf. 1993. Cladistics: What‘s in a word? Cladistics 9:183— 199. Korneyev, V. A. 1985. Significance of the periphallic complex of flies of family , Tephritidae for species diagnosis and improving'the classification of the family. In Systematics of Diptera (Insecta). Ecological and morphological principles. (0. A. Skarlato, ed). Oxomom Press, New Deli, 185 pp. Komeyev,- V. A. 1986. Fruit flies of the tribe Terelliini (Diptera, Tephritidae) in the USSR. Entomol. Rev. 65: 35—55. Korneyev, V. A. 1991. Tephritid flies of the genera allied to Euleia (Diptera, Tephritidae) of the USSR. fauna. Vestn. Zool. 3: 8—17. Lengyel, I, and l. R. Epstein. 1991. Modeling of Turing structures in the chlorite- iodide-malonic acid-starch reaction system. Science 251: 651—652. Maddison, D. R. 1991. The discovery and importance of multiple islands of most- parsimonious trees. Syst. Zool. 40:315—328. Maddison, D. R., M. Ruvolo, and D. L. Swofford. 1992. Geographic origins of human mitochondrial DNA: Phylogenetic evidence from control region sequences. Syst. Biol. 41:111—124. Maddison, W. P. 1993. Missing data versus missing characters in phylogenetic analysis. Syst. Biol. 42:576—581. Maddison, W. P., and D. R. Maddison. 1992. MacClade, version 3.0. Analysis of phylogeny and character evolution. Sinauer Associates, Inc., Sunderland, Massachusetts. Marquez, G. G. 1991. One hundred years of solitude. Harper and Row, New York. McAlpine, J. F. 1981a. Morphology and terminology—adults. Pages 9—63 In Manual of Nearctic Diptera, vol. 1 (J. F. McAlpine, B. V. Peterson, G. E. Shewell, H. J. Teskey, J. R. Vockeroth, and D. M. Wood, coords.). Res. Br., Agn'c. Can., Monogr. 27, Ottawa. 181 McAlpine, J. F. 1981b. Morgea freidbergi new species, a living sister-species of the fossil species M. mcalpinei, and a key to world genera of Pallopteridae (Diptera). Can. Ent. 113: 81—91. McAlpine, J. F. 1987. Lonchaeidae. Pages 791—797 In Manual of Nearctic Diptera, vol. 2 (J. F. McAlpine and D. M. Wood, coords). Res. Br., Agric. Can., Monogr. 32, Ottawa. McAlpine, J. F. 1989. Phylogeny and classification of the Muscamorpha. Pages 1,397—1,518 in Manual of Nearctic Diptera, vol. 3 (J. F. McAlpine and D. M. Wood, coords). Res. Br., Agric. Can., Monogr. 32, Ottawa. McPheron, B. A. and H.-Y. Han. submitted. Phylogenetic analysis of North American Rhagoletis (Diptera: Tephritidae) and related genera using mitochondrial DNA sequence data. Molecular Phylogenetics and Evolution. Meinhardt, H. 1982. Models of biological pattern formation. Academic Press, London. xi + 230 pp. Merz, B. 1994. Diptera: Tephritidae. Insecta Helvetica Fauna. Vol. 10. 198 pp. Microsoft Corporation. 1992—1993. EXCEL version 4.0. Microsoft Corp., Redmond, WA. Miles, D. B. and A. E. Dunham. Historical perspectives in ecology and evolutionary biology: the use of phylogenetic comparative analyses. Annu. Rev. Ecol. Syst. 24 :5 8 7- 6 1 9. Ming, Y. 1989. A revision of the genus Eurosta Loew, with a scanning electron microscope study of taxonomic characters (Diptera: Tephritidae). M.S. thesis, Washington State University, Pullman. 190 pp. Ming, Y. 1996. Characterization of the ribosomal DNA of the genus Rhagoletis (Diptera: Tephritidae). Ph. D. dissertation, Michigan State University. Munro, H. K. 1947. African Trypetidae (Diptera). Mem. Ent. Soc. Sth. Afr. 1. 284 pp. Munro, H. K. 1984. A taxonomic treatise on the Dacidae (T ephritoidea, Diptera) of Africa. South Afr. Dep. Agric. Water Suppl., Entomol. Mem. 61. ix + 313 pp. Murray, J. D. 1981. A pre-pattern formation mechanism for animal coat markings. J. theor. Biol. 88: 161—199. Murray, J. D. 1988. How the leopard gets its spots. Scientific American 258: 80—87. Murray, M. A., M. Schubiger, and J. Palka. 1984. Neuron differentiation and axon growth in the developing wing of Drosophila melanogester. Dev. Biol. 104: 259— 273. Nation, J. L. 1972. Courtship behavior and evidence for a sex attractant in the male Caribbean fruit fly, Anastrepha suspense. Ann. Entomol. Soc. Am. 65: 1,364— 1,367. 182 Nation, J. L. 1981. Sex-specific glands in tephritid fruit flies of the genera Anastrepha, Ceratitis. Dacus and Rhagoletis (Diptera: Tephritidae). Int. J. Insect Morphol. & Embryol. 10: 121—129. Nation, J. L. 1985. Respiratory systems. Pages 185—225 in Fundamentals of insect physiology (M. S. Blum, ed.). John Wiley and Sons, New York. xiv + 598 pp. Neff, N. A. 1986. A rational basis for a priori character weighting. Syst. Zool. 35:1 1 0—1 23. Nijhout, H. F. 1985. The developmental physiology of color patterns in Lepidoptera. Adv. Insect Physiol. 18: 181—247. Nijhout, H. F. 1991. The development and evolution of butterfly wing patterns. Smithsonian Institution Press, Washington, D. C., xvi + 297 pp. Nixon, K. C., and O. D. Wheeler. 1990. An amplification of the phylogenetic species concept. Cladistics 62211—223. Norrbom, A. L. 1989. The status of Urophora acuticarnis and U. sabroskyi (Diptera: Tephritidae). Ent. News 100(20): 59—66. Norrbom, A. L. 1990. Notes on Zonosemata Benjamin (Diptera: Tephritidae) and the status'of Cryptodacus scutellatus Hendel (= 2. ice Steyskal syn. n.). Ann. Naturhist. .Mus. Wien 91:53—55. Norrbom, A. L. 1993. New species and phylogenetic analysis of Euaresta Loew (Diptera: Tephritidae), with a key to the species from the Americas south of Mexico. Proc. Entomol. Soc. Wash. 95: 195—209. Norrbom, A. L. 1993. New synonymy of Epochra Loew with Euphranta (Rhacochlaena Loew) (Diptera: Tephritidae) and description of a new species from Mexico. Proc. Entomol. Soc. Wash. 95:189—194. Norrbom, A. L. 1994. New species and phylogenetic analysis of Cryptodacus, Haywardina, and Rhagoletotrypeta (Diptera: Tephritidae). Insecta Mundi 8(1—2): 37—65. Norrbom, A. L., and K. C. Kim. 1988. Revision of the schausi group of Anastrepha Schiner (Diptera: Tephritidae), with a discussion of the terminology of the female terminalia in the Tephritoidea. Ann. Entomol. Soc. Am. 81(2): 164—173 Norrbom, A. L., Y. Ming, and V. Hemandez-Ortiz. 1988. A revision of the genus Oedicarena Loew (Dipter: Tephritidae). Fol. Entomol. Mex. 75: 93—117. Novak, J. A. 1974. A taxonomic revision of Dioxyna and Paroxyna (Diptera: Tephritidae) for America north of Mexico. Melandaria 16. 53 pp. Oreskes, N., K. Shrader-Frenchette, and K. Belitz. 1994. Verification, validation, and confirmation of numerical models in the earth sciences. Science 263: 641—646. Palka, J., M. A. Malone, R. L. Ellison, (and D. J. Wigston. 1986. Central projections of identified Drosophila sensory neurons in relation to their time of development. J. Neurosci. 6: 1,822—1,830. 183 Pickett, A. D.’ 1937. Studies on the genus Rhagoletis (Trypetidae) with special reference to Rhagoletis pomonella (Walsh). Can. J. Res., Sec. D, 15: 53—75. Pimentel, R. A., and R. Riggins. 1987. The nature of cladistic data. Cladistics 3:201— 209. Platnick, N. I., C. E. Griswold, and J. A. Coddington. 1991. On missing entries in cladistic analysis. Cladistics 7:337—343. Pool, R. 1991. Did Turing discover how the leopard got its spots? Science 251: 627. Prokopy, R. J., and J. Hendrichs. 1979. Mating behavior of Ceratitis capitata on a field-caged host tree. Ann. Entomol. Soc. Am. 72: 642—648. Richter, V. A. 1970. Chapter 62. Family Tephritidae (Trypetidae). In Bienko, G. Ya. Bei (ed.) Keys to the insects of the European part of the U.S.S.R. Opred. Faune S.S.S.R. 103(v):132—172 [in Russian; English trans. pp. 212—276]. Shiraki, T. 1933. A systematic study of the Trypetidae in the Japanese Empire. Mem. Fac. Sci. and Agri., Taihoku Imp. Univ., Formosa, Japan. Vol. VIII, 509 pp. + 14 pls. Shiraki, T. 1968. Fruit flies of the Ryukyu Islands. U. S. National Museum Bull. 263. Sinclair, B. J., J. M. Cumming, and D. M. Wood. 1994. Homology and phylogenetic implications of male genitalia in Diptera—lower Brachycera. Entomol. Scand. 24: 407—432. Smith, J. J. and G. L. Bush. in review Phylogeny of the genus Rhagoletis (Diptera: Tephritidae) inferred from DNA sequences of mitochondrial cytochrome oxidase II. Steyskal, G. C. 1958. Notes on the Richardiidae, with a review of the species known to occur in the United States (Diptera, Acalyptratae). Ann. Entomol. Soc. Am. 51: 30 2—31 0 . Steyskal, G. C. 1979. Taxonomic studies on fruit flies of the genus Urophora (Diptera: Tephritidae). Special Publ., Entomol. Soc. Wash., 61 pp. Steyskal, G. C. 1984. A synoptic revision of the genus Aciurina Curran, 1932 (Diptera: Tephritidae). Proc. Entomol. Soc. Wash. 86(3): 582—598. Steyskal, G. C. 1986. Taxonomy of the adults of the genus Streuzie Robineau-Desvoidy (Diptera, Tephritidae). Insecta Mundi 1(3):101—117. Steyskal, G. C. 1987a. Platystomafidae. Pages 809—812 In Manual of Nearctic Diptera, vol. 2 (J. F. McAlpine, B. V. Peterson, G. E. Shewell, H. J. Teskey, J. R. Vockeroth, and D. M. Wood, coords). Res. Br., Agric. Can., Monogr. 28, Ottawa. Steyskal, G. C. 1987b. Pyrgotidae. Pages 813—816 In Manual of Nearctic Diptera, vol. 2 (J. F. McAlpine, B. V. Peterson, G. E. Shewell, H. J. Teskey, J. R. Vockeroth, and D. M. Wood, coords). Res. Br., Agric. Can., Monogr. 28, Ottawa. 184 Stoffolano, J. G. and Yin, L. R. S. 1987. Structure and function of the ovipositor and associated sensilla of the apple maggot, Rhagoletis pomonella (Walsh) (Diptera: Tephritidae). Int. J. Insect Morphol. Embryol. 16:41—69 Stoltzfus, W. B. 1977. The taxonomy and biology of Eutreta (Diptera: Tephritidae). Iowa State J. Res. 51(4): 369—438 Stoltzfus, W. B. 1988. The taxonomy and biology of Strauzia (Diptera: Tephritidae). J. Iowa Acad. Sci. 95(4): 117—126. Swofford, D. L. 1993. PAUP: Phylogenetic analysis using parsimony, version 3.1.1. Illinois Natural History Survey, Champaign. van Bemmelen, J. F. 1917. The colour pattern on Diptera wings. Proc. Koninklijke Akademie van Wetenschappen te Amsterdam 19: 1,141—1,156. Waddington, C. H. 1941. The genetic control of wing development in Drosophila. J. Genetics 41: 75—137. Wang, X-J. and M-2. Zhao. 1989. Notes on the genus Dacus Fabricius in China with descriptions of five new species (Diptera: Tephritidae). Acta Zooltaxonomica Sinica 14(2):209-—219. Watrous, L. E., and O. D. Wheeler. 1981. The outgroup comparison method of character analysis. Syst. Zool. 30:1—11. White, I. M. 1988. Tephritid flies—Diptera: Tephritidae, pp. 1—134. In P. C. Barnard and R. R. Askew [eds.], Handbooks for the identification of British insects, vol. 10, pt. 5a. Royal Entomological Society of London. White, I. M. and M. M. Elson-Harris. 1992. Fruit flies of economic significance: Their identification and bionomics. CAB. International, Oxon, UK. xii + 601 pp. Wiegmann, B. M., C. Mitter, and F. C. Thompson. 1993. Evolutionary origin of the Cyclorrhapha (Diptera): Tests of alternative morphological hypotheses. Cladistics 9: 41—81. Wigglesworth, V. B. 1950. A new method for injecting the tracheae and tracheoles of insects. Quart. J. Micro. Sci. 91(2):217—224. Wigglesworth, V. B. 1963. A further function of the air sacs in some insects. Nature, Lond. 198:106. Wiley, E. O. 1981. Phylogenetics. The theory and practice of phylogenetic systematics. John Wiley and Sons, New York. xv + 439 pp. Wood, D. M. 1991. Homology and phylogenetic implications of male genitalia in Diptera. The ground plan. Pp. 255—284 In Weismann, Orsza'gh and Pant: Proc. Second Int. Congress Dipterology. The Hague. Yeates, D. 1992. Why remove autapomorphies? Cladistics 8:387—389.