Anastrepha Main | Toxotrypana Main | Tephritidae Main | Diptera Home | SEL Home

Phylogeny of Anastrepha and Toxotrypana

The following text is copied from:
Norrbom, A.L., R.A. Zucchi & V. Hernández-Ortiz. 1999. Phylogeny of the genera Anastrepha and Toxtrypana (Trypetinae: Toxotrypanini) based on morphology, p. 299-342. In M. Aluja & A. L. Norrbom, eds., Fruit flies (Tephritidae): Phylogeny and evolution of behavior. CRC Press, Boca Raton. [16] + 944 p.

For additional analysis see:
McPheron, B. A., H.-Y. Han, J. G. Silva & A. L. Norrbom. 1999. Phylogeny of the genera Anastrepha and Toxotrypana (Trypetinae: Toxotrypanini) based upon 16S rRNA mitochondrial DNA sequences, p. 343-361. In M. Aluja & A. L. Norrbom, eds., Fruit flies (Tephritidae): Phylogeny and evolution of behavior. CRC Press, Boca Raton. [16] + 944 p.

CONTENTS

1. Introduction

2. Morphological Characters Useful in Anastrepha and Toxotrypana

3. Relationships of Anastrepha and Toxotrypana

4. Relationships within Toxotrypana

5. Intrageneric Classification and Relationships within Anastrepha

5.1. Relationships Among Species Groups of Anastrepha

5.2. Species Groups of Anastrepha

5.2.1. The cryptostrepha Group

5.2.2. The daciformis Group

5.2.3. The dentata Group

5.2.4. The benjamini Group

5.2.5. The robusta Group

5.2.6. The schausi Group

5.2.7. The punctata Group

5.2.8. The leptozona Group

5.2.9. The mucronota Group

5.2.10. The grandis Group

5.2.11. The doryphoros Group

5.2.12. The spatulata Group

5.2.13. The ramosa Group

5.2.14. The pseudoparallela Group

5.2.15. The serpentina Group

5.2.16. The striata group

5.2.17. The fraterculus Group

5.2.18. Unplaced Species

6. Host Plant Relationships

7. Conclusions and Future Research Needs

8. Acknowledgments

9. Literature Cited

1. INTRODUCTION

Anastrepha Schiner is the largest and most economically important genus of Tephritidae in the Americas, including major pest species such as the Mexican, Caribbean, and South American fruit flies. The closely related genus Toxotrypana Gerstaecker also contains economically important species, including the papaya fruit fly. Despite their economic status, little has been published about the phylogeny of Anastrepha and Toxotrypana, as is true for most groups of Tephritidae. In this chapter, we analyze the relationships among the species of these two genera based on morphological character data, including new as well as previously published information. We also summarize host plant data at the level of plant family and discuss several trends in host plant relationships.

Other than the phylogenetic analysis of the 13 species of the Anastrepha daciformis group by Norrbom (1998), there has been no rigorous cladistic analysis of Anastrepha or Toxotrypana using computer parsimony programs such as Hennig86 or PAUP. Although we have analyzed character polarities (Table 2) and present some preliminary phylogenetic hypotheses in this chapter, we have not conducted a rigorous analysis because we doubt that there would be much resolution beyond the species group level at this time, and even some of the species groups would not be supported because of missing data or variable characters. Compared with the number of species, relatively few characters have been discovered that appear to be useful for phylogenetic analysis in these genera. Many of the taxonomically useful characters, such as aculeus length, intergrade to such an extent that dividing them into character states is problematic. For most of the morphological characters we have studied, the apomorphic state occurs in a relatively small number of species. We have found few apomorphies above this level. Thus our discussion focuses on the definition of the species groups. We indicate whether diagnostic character states are apomorphic, and if they are not unique, in what other species they are present. Some judgement regarding which character states are likely to be homoplastic is thus implied, but such hypotheses need to be further tested by more rigorous analysis and additional character data.

2. MORPHOLOGICAL CHARACTERS USEFUL IN ANASTREPHA AND TOXOTRYPANA

Color patterns on the body are useful characters in Toxotrypana and many Anastrepha species (Figs. 1-2). The markings on the thorax are the most important taxonomically, although in some species there are markings on the head (e.g., some species of the A. daciformis group and males of some species of the A. schausi group) or the abdomen (e.g., A. serpentina group). Anastrepha species have a pair of sublateral stripes of the xanthine type, and frequently an unpaired medial one, although in generally pale colored species they are often not well contrasted and may be difficult to see, especially in dried, preserved specimens. But they are obvious in generally dark bodied species (Fig. 1B, J). In the A. fraterculus group, in which most of the body is yellow or orange, the subscutellum and/or the mediotergite may have a characteristic dark brown lateral stripe or spot (Fig. 2E-F). These markings vary intraspecifically in some species, but are nonetheless very useful taxonomically.

Microtrichia patterns, especially on the scutum, are often useful characters in Anastrepha (Norrbom 1985). As in most Tephritidae, the microtrichia are extremely fine and can be differentiated well only with a scanning electron microscope (see Fig. 33.2C). Under the light microscope, where they are present the body surface looks duller or matte if they are very dense. Where there are no microtrichia, the cuticle appears shiny. The scutal microtrichia should not be confused with the scutal setulae, which are much larger and have alveoli, or sockets. In Anastrepha the scutum is frequently entirely microtrichose, but in many species it is bare, microtrichose only laterally and posteriorly, or sometimes (e.g., in the striata and doryphoros groups) there are nonmicrotrichose stripes. This character is subtle and difficult to see (particularly on specimens in alcohol or those from McPhail traps that often are covered with fine particulate matter), but is very useful taxonomically. Its appearance may vary depending upon the angle of observation, and it is best viewed from an oblique anterodorsal angle.

The facial carina (= clypeal ridge of Stone 1942) is produced in some Anastrepha species (see Norrbom 1997). This is best seen in lateral view. In most species it is straight or concave in profile (Fig. 4A-B), but the medial or dorsal parts may project giving a convex appearance (Fig. 4C, E-F).

Most species of Anastrepha have a similar wing pattern (Fig. 5E, 6A), consisting of three bands that have been termed the C-band, S-band, and V-band (Stone 1942, Steyskal 1977b). The C-band, or costal band, runs from the wing base along the anterior margin to the apex of vein R₁, filling cells bc, c, sc, and the bases of r₁, r₂₊₃, and br. It is often fainter in cells bc and c and/or is darker in the pterostigma. The S-band, which is somewhat S-shaped if viewed from the wing apex, runs from cell bcu, obliquely across R-M to the anterior wing margin, and then follows the margin to beyond the apex of vein R₄₊₅. The V-band forms an inverted V, with its apex on or near vein R₄₊₅; the subapical band (covering DM-Cu) forms the proximal arm, and the posterior apical band (crossing cell m) forms the distal arm. The wing patterns of about 95% of the species of Anastrepha have some variation of this pattern or are easily derived from it. The three bands may be separated (Fig. 6A) or connected (Fig. 5E); usually they touch along vein R₄₊₅ when they are connected, although they occasionally connect elsewhere. Parts of some bands may be absent, especially the apex and/or the distal arm of the V-band.

A few species of Anastrepha have diffuse wing patterns, for example, A. doryphoros and obscura, but in most cases species closely related to them have typical or intermediate patterns. The male of A. bellicauda also has a very diffuse pattern (Fig. 5J), but the recently discovered female has a more typical Anastrepha type pattern (Fig. 5I). A few other species of Anastrepha and all of the species of Toxotrypana have a wasp mimicry wing pattern (Fig. 5A, D), with only a long costal band that is not interrupted at the apex of vein R₁, and an infuscated area in cell bcu and along vein A₁+Cu₂. There are species with intermediate patterns in the A. daciformis and grandis groups (Fig. 5C, 6C).

The main useful characters of the male genitalia (Fig. 8) in Anastrepha and Toxotrypana are the length of the phallus, which is correlated with the length of the female genitalia, and the shape of the surstyli, especially the lateral surstylus (Fig. 8-10).

The female genitalia (Fig. 12) of Anastrepha and Toxotrypana have an enlarged dorsobasal area of the eversible membrane that bears moderately to greatly enlarged scales or teeth (Fig. 12A, C, 13-15). These large teeth have been called "the rasper" although there is no evidence regarding their function and use of that term is not recommended.

The shape, dentition, and length of the aculeus tip (Fig. 16-17) are extremely variable in Anastrepha and along with aculeus length are among the most important taxonomic characters at the species level. The length of the aculeus tip has been defined as the distance from the apex of the inner margin of the sclerotized area on the ventral side of the aculeus (Fig. 12D, arrow) to the extreme apex. The former has sometimes been referred to as the opening of the cloaca or oviduct, but the cloacal opening is actually membranous and difficult to see, and is slightly more basal.

Regarding the immature stages, eggs have been described to some extent for T. curvicauda and 25 species of Anastrepha (see Table 1). There are some phylogenetically useful characters in their gross morphology (Fig. 3), particularly the presence of lobes in some species. Other characters, involving surface sculpture and vestiture, have been described in a few species , but need to be studied in greater detail (e.g., as by Murillo & Jirón 1994 and Selivon & Perondini, in press) and in more species. The lobes on the eggs, at least in A. obliqua, have a respiratory function (Seín 1933, Murillo & Jirón 1994). Their presence may be a relatively plastic character (i.e., easily evolved and subject to homoplasy). There are species with similar lobes (although often with the micropyle at the apex of the lobe) and other species without them within each of the genera Aciurina, Chaetorellia, and Rhagoletis, and species with lobes are also known in Craspedoxantha and Paracantha (Tauber & Tauber 1967, Freidberg 1985, White & Marquardt 1989, Headrick & Goeden 1990, 1993, Frías et al. 1993).

The larval stages of Anastrepha and Toxotrypana are also poorly known. First and/or second instars have been described only for A. bistrigata, A. grandis, and A. ludens (Steck & Malavasi 1988, Steck & Wharton 1988, Carroll & Wharton 1989). Third instars have been described to some extent for T. curvicauda and 16 species of Anastrepha (Heppner 1986, Norrbom 1985, Steck et al. 1990, White & Elson-Harris 1992). Most of the taxonomically useful characters are difficult to interpret phylogenetically. Most, such as the number of oral ridges, number of tubules of the anterior spiracle, or presence of spinules on the thorax and abdomen, are meristic and/or overlap considerably among the species and are difficult to divide into character states for phylogenetic analysis.

3. RELATIONSHIPS OF ANASTREPHA AND TOXOTRYPANA

Anastrepha and Toxotrypana belong to the tribe Toxotrypanini, which otherwise includes only the genus Hexachaeta Loew. All three genera are primarily Neotropical, although a few species of each genus extend slightly into the Nearctic Region.

Hexachaeta, which is currently being revised by Hernández-Ortiz, includes about 25 described and numerous undescribed species. These species comprise several groups that are well-defined by morphological characters. Hexachaeta’s affinities have been enigmatic (Foote et al. 1993). It has been included in the Acanthonevrini on the basis of having three pairs of scutellar setae (Foote 1967), but Hancock (1986) and Korneyev (1994) rejected this hypothesis because the aculeus tip is completely fused and there are small denticles on the spermathecae. The latter author placed Hexachaeta in a monotypic tribe, Hexachaetini. Hexachaeta was tentatively included in the Toxotrypanini by Norrbom et al. (1999a) based mainly on results of a molecular study by Han & McPheron (1997). Although there is some morphological resemblance between Hexachaeta and the other two genera, to date no morphological synapomorphies have been proposed to support or contradict their relationship. Some Hexachaeta species, especially the colombiana group, somewhat resemble Anastrepha in body markings and wing pattern. Most species of Hexachaeta have the apical extension of cell bcu relatively large as in both Anastrepha and Toxotrypana, but this character state is not unique to the Toxotrypanini. The ocellar bristles are poorly developed in the species of the H. socialis and amabilis groups, as in all species of Toxotrypana and all but one or two species of Anastrepha. This is an apomorphic character state, but it occurs within various other higher groups of Tephritidae (Hernández-Ortiz, unpubl.). The male genitalia in Hexachaeta include very long, slender surstyli. In the female, the aculeus may be simple, multilobed, or lobed and serrate (Lima 1953a, b, 1954).

Although Toxotrypana was long placed in the Dacina (Loew 1873, Hardy 1955), there is strong morphological evidence that it forms a monophyletic group with Anastrepha. Snow (1895) was the first to suggest this relationship, but his comments were apparently unknown to, or were ignored by, subsequent taxonomists who classified Toxotrypana. Only recently was the relationship of Anastrepha and Toxotrypana convincingly demonstrated. Kitto (1983) and Sarma et al. (1987) found T. curvicauda and A. suspensa (Loew) to be similar in an immunological analysis, and suggested that what had been considered strong morphological similarity of Toxotrypana and the Dacina was due to convergence. Re-examinations by Norrbom (1985), Hancock (1986), and Norrbom & Foote (1989) demonstrated that the resemblance of Toxotrypana and the Dacina actually is superficial (e.g., reduced chaetotaxy, but in different ways in each group). Morphology also strongly indicates the monophyly of Toxotrypana and Anastrepha, including the following synapomorphies: 1) the eversible membrane of the female is enlarged basally and bears a group of enlarged dorsal teeth (the teeth are secondarily reduced in some species of the A. dentata group, but still larger than in this area in other tephritids, and the basal area is still enlarged) (Fig. 12-15); 2) the base of the oviscape of the female has lateral, flange-like lobes (smaller in Toxotrypana) (Fig. 12A-B); 3) the male glans is weakly sclerotized medially and has a T-shaped, somewhat hook-like apical sclerite (Fig. 11); 4) the surstyli are relatively short, the lateral surstylus without anterior or posterior lobes (Fig. 8-10); and 5) vein M is anteriorly curved in the distal half of its last segment (distal to DM-Cu) (Fig. 5-6). Other characters of uncertain polarity, such as the strongly sclerotized secondary connection of the subepandrial sclerite, the number, size and arrangement of the sensilla of the aculeus tip (3 pairs, none extended beyond lateral margin), and the posterior location of the dorsocentral seta, are similar in Toxotrypana and Anastrepha. These similarities do not contradict the hypothesis that Anastrepha and Toxotrypana form a monophyletic group, and some or all may be additional synapomorphies.

Autapomorphies indicating the monophyly of Toxotrypana include: 1) vein R₂₊₃ with 3 sharp bends, often with spur veins arising from them (Fig. 5A); 2) male wing with costal setulae stout (sometimes intraspecifically variable) (Fig. 5A); 3) wing pattern consisting of only broad complete costal band and faint streak over cell bcu (Fig. 5A) (this also occurs, probably due to convergence, in some species of the Anastrepha daciformis and grandis species groups); 4) subapical bend in vein Sc weak; 5) scutum with a medial longitudinal depression; 6) many setae reduced (ocellar, frontal, orbital, postpronotal, acrostichal, dorsocentral, presutural supra-alar, anepisternal, katepisternal, and basal scutellar setae very small and weak or absent; each is at least sometimes present in at least one species); 7) abdomen petiolate (Fig. 2C); and 8) body yellow with dark markings, or predominantly dark (Fig. 2A-C).

Although there is strong morphological evidence for the monophyly of Toxotrypana, and for Toxotrypana + Anastrepha, the same cannot be said for Anastrepha alone. The apical curvature of the medial vein (e.g., Fig. 5D, 6) was long considered a diagnostic character for the genus, but in a few species (e.g., Fig. 5B, J, K) the curvature is weak and no stronger than what occurs within Toxotrypana. Within some species of Toxotrypana (e.g., T. nigra) (Fig. 5A), this vein meets the costa without an obvious angle, thus the range in this character overlaps in the two genera. The anterior bend in the distal half of the distal segment of the medial vein is a synapomorphy for both genera, but some Anastrepha species cannot be differentiated from Toxotrypana on the basis of having this vein more strongly curved. A possible synapomorphy for Anastrepha suggested by Norrbom (1985) is the typical wing pattern, consisting of C-, S- and V-bands, and occurring in more than 90% of the species. This assumes that the atypical patterns found in a few species are secondary modifications, as appears to have occurred within the A. daciformis group in which there is a transition from the typical Anastrepha pattern to the wasp mimicry pattern (Norrbom 1998). However, because the wasp mimicry pattern has evolved at least twice within Anastrepha (in A. aberrans and within the daciformis group), the hypothesis that this pattern also could have evolved in Toxotrypana from the Anastrepha pattern should not be considered unlikely. Recent molecular studies (McPheron et al., this volume) in fact suggest that some species of Anastrepha may be more closely related to Toxotrypana, or in other words, that Anastrepha is not monophyletic without including Toxotrypana. This would cause a nomenclatural problem, as the latter name has priority, but if these genera eventually need to be synonymized, Anastrepha should be used to conserve usage for the major pest species it includes. Because the monophyly of Anastrepha is in doubt, Toxotrypana should be included in any analysis of relationships within Anastrepha. We have used Hexachaeta as the outgroup for the analyses discussed in this chapter.

4. RELATIONSHIPS WITHIN TOXOTRYPANA

Toxotrypana includes seven described and at least six undescribed species (Table 4; Norrbom & Zucchi, in prep.). There are two probably monophyletic species groups. The species of the curvicauda group have extremely long, strongly curved female terminalia, as in the best known species, T. curvicauda. This is an apomorphic character, although it occurs, probably via convergence, in an undescribed Anastrepha species. The other species group within Toxotrypana includes at least four species, all undescribed, that have shorter, straight female terminalia, and the apical setae of male tergite 5 often short and stout.

Both species groups occur from Mexico to Central America, and following the Andes to Argentina and southern Brazil. T. curvicauda also occurs in the Antilles. The known host plants belong to milky latex bearing families: Asclepiadaceae, Caricaceae, and possibly Apocynaceae. The larvae feed on the developing seed tissues inside the thick husked fruits.

5. INFRAGENERIC CLASSIFICATION AND RELATIONSHIPS WITHIN ANASTREPHA

Anastrepha includes 197 currently recognized species (Table 4). Of these valid names, about ten are poorly recognized and may be synonyms; they were insufficiently described and were based on lost, damaged or male type specimens. On the other hand, we know of nearly 50 additional undescribed species, and there are undoubtedly many more yet to be discovered. Some of what were thought to be widespread species, such as A. fraterculus, hamata and pickeli (Steck 1991, Norrbom 1985, Canal 1997), are now known to be cryptic species complexes. Resolution of these complexes and the likelihood that other complexes occur within Anastrepha will further increase the total number of species.

The first groupings of species now classified within Anastrepha were proposed by Stone (1939a, b), who recognized two separate genera that were subsequently synonymized with Anastrepha by Steyskal (1977a): Pseudodacus (including 4 species now in the daciformis group) and Lucumaphila (including 11 species, 7 now in the dentata group). Acrotoxa, which was proposed by Loew (1873) without knowledge of Schiner’s genus Anastrepha, and Phobema, proposed by Aldrich (1925) as a monotypic genus, are also synonyms of Anastrepha. Shaw (1962) first recognized the spatulata group for four species, two of which are now considered synonyms of A. spatulata. Korytkowski & Ojeda (1968) recognized ten numbered groups for 35 species they studied from Peru. Steyskal (1977b) recognized four additional species groups, the benjamini, grandis, punctata and serpentina groups. Zucchi (1977) proposed the fraterculus species group, and Norrbom (1985) recognized an additional 12 species groups or subgroups, but only some of the results of these dissertations have been formally published and widely circulated. Norrbom & Kim (1988b) further modified the intrageneric classification and provided a checklist of the then recognized 180 Anastrepha species divided into 17 species groups, but they did not explain the basis for this classification. Norrbom (1991, 1998) revised the grandis and daciformis groups and added additional species. Norrbom (1997) suggested that the benjamini group (sensu Steyskal 1977b) was not monophyletic and transferred some species from it, as well as all of the species of the chiclayae group, to the pseudoparallela group. In a publication that came to our attention just prior to this chapter going to press, Tigrero (1998) grouped 31 species from Ecuador into 17 species groups. He did not explain the basis for this classification, however, and we have not had time to evaluate it thoroughly.

Other than that by Norrbom (1998) for the daciformis group, there has been no rigorous cladistic analysis of Anastrepha based on morphological characters using computer software. Although the relationships among most of the species groups are poorly understood, many of the species groups that have been recognized appear to be monophyletic (the status of others needs further analysis). These species groups and their included species are listed in Table 4. Nomenclatural and distributional data for most species are listed by Norrbom et al. (1999b). The limited previous analysis of relationships among the species groups of Anastrepha used Toxotrypana as the outgroup (Norrbom 1985), but new molecular data (McPheron et al., this volume) suggest that Anastrepha may be paraphyletic without including Toxotrypana. Our present analysis therefore includes Toxotrypana as part of the ingroup, and uses Hexachaeta as the outgroup. Table 2 lists most of the characters that are discussed and indicates their polarities, and Table 3 shows the distributions of their states. A preliminary cladogram indicating possible relationships is shown in Fig. 7.

5.1. RELATIONSHIPS AMONG SPECIES GROUPS OF ANASTREPHA

Little morphological evidence has been discovered to indicate the relationships among the species groups of Anastrepha and Toxotrypana. The A. cryptostrepha group may be the most primitive clade (see discussion of that group). There is considerable evidence that the A. daciformis and dentata groups are sister groups, including the following synapomorphies: 1) phallus short, less than 1.25 mm long, and glans absent (Fig. 9E); 2) aculeus extremely slender (Fig. 16C-D), less than 0.05 mm wide except at base (reduction in aculeus width has occurred in certain other Anastrepha species, but whether this is convergence or a synapomorphy for some of these species and the daciformis + dentata groups remains uncertain); and 3) spermathecae weakly sclerotized. Norrbom (1998) suggested another possible synapomorphy: Third instar larva with hairs of hind spiracle relatively short, but this character is known for only one species in each group, A. pallens and A. sagittata (Baker et al. 1944, Phillips 1946), and these hairs are also short in Toxotrypana, and relatively short in A. interrupta (spatulata group) and A. limae (pseudoparallela group) (Baker et al. 1944, Steck & Wharton 1988), so this character may be homoplasious or have evolved at a lower level. The daciformis and dentata groups share one additional apomorphy: Lateral surstylus very short and rounded, barely extended beyond the prensisetae (Fig. 9F, H)), and subepandrial sclerite relatively posterior in position, in lateral view its apex usually at posterior margin of epandrium (Fig. 9E, G). But here again, the lateral surstylus is also rather short in Toxotrypana, the A. robusta group (especially in A. binodosa), and to a lesser extent in the A. punctata, leptozona and schausi groups, so it could possibly be a synapomorphy for some or all of these groups in addition to the A. daciformis and dentata groups.

The A. punctata group resembles some species of the A. cryptostrepha and leptozona groups in lateral surstylus shape (Fig. 9I-J), but the surstylus is oriented obliquely rather than being broadest in the longitudinal or transverse directions. Vein M is usually weakly curved apically (plesiomorphy) in the punctata and schausi groups, as in the cryptostrepha group (Fig. 5B, I-K, but this state occurs in a few species in other species groups as well (e.g., A. superflua of the benjamini group).

A large clade may be formed by the A. pseudoparallela + spatulata + ramosa + doryphoros + grandis + serpentina + striata + fraterculus groups. In these groups the male proctiger has strong lateral creases, with the sclerotized area divided into 3 parts (Fig. 10E, I, K, N). In other Anastrepha species groups and Toxotrypana it is weakly or not creased and the sclerotized area is usually continuous (Fig. 8B, 9A, C, E, G).

5.2. SPECIES GROUPS OF ANASTREPHA

5.2.1. The cryptostrepha Group

The following character states suggest that this group may be monophyletic: 1) male lateral surstylus very broad in lateral view (Fig. 9C-D), not transversely flattened (polarity uncertain); 2) scutum mostly or entirely bare of microtrichia (apomorphic, but occurs sporadically outside this group); 3) aculeus tip short, nonserrate, and nearly round in cross-section, approximately as broad in lateral view as in ventral view (Fig. 16A-B) (apomorphic); and 4) vein M weakly curved apically (Fig. 5B) (plesiomorphic, occurs occasionally in other groups). This may be the most primitive species group of Anastrepha if the first character is plesiomorphic (i.e., if the more transversely flattened condition found in other Anastrepha species groups is apomorphic), but the interpretation of this character is equivocal as there is no distinct anterior or posterior lobe on the lateral surstylus in the cryptostrepha group as there is in Hexachaeta and most Trypetinae.

Within the cryptostrepha group, A. cryptostrepha and cordata share a unique apomorphy of the male genitalia: The proctiger is sclerotized dorsally. The scutal markings of these species, although much less extensive in A. cryptostrepha, are another possible synapomorphy: A. cryptostrepha usually has a moderate brown band along the posterior margin (similar to Fig. 1C), whereas A. cordata has extensive dark brown markings, including along the posterior margin (Fig. 1A). Anastrepha margarita, panamensis and zeteki appear to comprise another group of closely related species. They share the following probable synapomorphies: scutellum without microtrichia dorsally; aculeus very slender, less than 0.06 mm wide at midlength (reduction in aculeus width occurs in various other species groups); and lateral surstylus relatively long and acute.

Whether these two clades plus A. tripunctata form a monophyletic group is less certain (i.e., it is possible that the cryptostrepha group as a whole is paraphyletic). Anastrepha tripunctata possesses a possibly plesiomorphic character state, ocellar seta strong (Fig. 4A), that suggests it could be the sister taxon to the rest of Anastrepha + Toxotrypana. The head shape of A. tripunctata, which is as long as high, was also interpreted as plesiomorphic by Norrbom (1985), but this is probably apomorphic based on outgroup comparison with Hexachaeta. There are also two undescribed species from Mexico that appear to be related to A. tripunctata based on their scutellar markings and wing pattern; at least one of them has a well developed ocellar seta (C. Estrada, pers. comm.). Because the size of the ocellar seta is variable within Hexachaeta, the polarity of this character within Anastrepha is uncertain; the well developed seta could be a synapomorphy for A. tripunctata and the new species from Mexico, or the reduced seta could be a synapomorphy for all other Anastrepha and Toxotrypana species (the latter hypothesis requires that there is homoplasy in characters 2 and 3 above, but given their variability across the genus this does not seem unrealistic). Study of the new Mexican species and analysis of Hexachaeta to determine the ground plan condition of this character in that genus may help to resolve the relationships of A. tripunctata.

The cryptostrepha group is predominantly distributed from Mexico to Venezuela, although A. cryptostrepha is known only from Ecuador and Peru. The known host plants of the group (for 3 of the 8 spp.) are Apocynaceae and Sapotaceae.

5.2.2. The daciformis Group

This is probably the most clearly monophyletic species group of Anastrepha, as indicated by the following unique apomorphies: 1) scutellum bicolored, with at least basal third on sides and dorsum distinctly darker than apex, and with basal scutellar seta within darker area (Fig. 1B), except usually in A. avispa (scutellar markings are present on only a few other Anastrepha species, and in different patterns, suggesting they are not homologous); 2) eversible membrane with unique pattern of dorsobasal teeth (Fig. 13B), small and weakly sclerotized except for medially interrupted apical row of large, strongly sclerotized, hook-like teeth; 3) spermathecae membranous; and, 4) phallus extremely short, less than 0.30 mm long (Fig. 9E). Relationships among the species of the daciformis group were analyzed by Norrbom (1998) (see Phylogeny of Anastrepha daciformis group). The species with wasp mimicry wing patterns form a monophyletic group, including the daciformis complex (A. castanea, daciformis, and katiyari) and the macrura complex (A. aquila, avispa, bicolor, macrura, and zucchii), which together are the sister group of A. pallens. The relationships of the remaining four species (A. antilliensis, murrayi, maculata, and stonei) are less resolved. They either form a monophyletic group that is the sister group of the above group, or they arise as basal clades in the following order: antilliensis, murrayi, and maculata + stonei. Anastrepha nambacoli Tigrero (1992) is a synonym of A. macrura. Norrbom (1998) examined the holotype of nambacoli and reported it under macrura, not realizing the former name was available.

The daciformis group is widespread. A. antilliensis, murrayi, maculata and stonei are endemic to the Antilles. A. pallens is Mesoamerican, the daciformis complex is restricted to South America, and the macrura complex includes three Mesoamerican and two South American species. The known host plants (for 3 of the 13 spp.) are Sapotaceae. The larvae feed on the seeds.

5.2.3. The dentata Group

Two synapomorphies indicate that this species group is monophyletic: 1) apex of aculeus poorly defined, extreme apex usually broader and sagittate (Fig. 16D) (unique apomorphy); and 2) dorsobasal teeth of eversible membrane all small and weak (Fig. 13C). Reduction in size and sclerotization of the dorsobasal teeth of the eversible membrane occurs in a few other species with slender aculei (e.g., hamadryas), but not as extensively as in this group; whether or not this is homoplasy or a synapomorphy of these species plus the dentata group is undetermined. This group is roughly equivalent to the genus Lucumaphila proposed by Stone (1939a), now considered a synonym of Anastrepha, although Steyskal (1977b) and Norrbom & Kim (1988b) removed four of the 11 species he included and added two others.

The dentata group is widespread, from southern Texas to Brazil, with one Antillean species. The known host plants (for only 3 spp.) are Sapotaceae. The larvae feed in the seeds only and tunnel out through the pulp of the fruit.

5.2.4. The benjamini Group

Two apomorphic character states, neither unique to this group, support its monophyly: 1) facial carina strongly produced medially and convex in lateral view (Fig. 4E) (apomorphic but also occurs in the pallidipennis complex of the pseudoparallela group, and rarely in other species; a few other species, such as A. atrox and several new species of the mucronota group (Fig. 4C), have the carina produced more dorsally (see Norrbom 1997)); and 2) female terminalia and male phallus very long, oviscape greater than 8.5 mm long (apomorphic, but occurs in other groups). This group may not be monophyletic, as the more compelling possible synapomorphy, the produced facial carina, apparently has evolved convergently within other species groups. At least A. gigantea and magna appear to be closely related, however (Norrbom 1997). A. benjamini has dense microtrichia around the lobe of cell bcu in both sexes, an unusual, autapomorphic character of unknown behavioral significance.

The benjamini group includes two species from Panama, one from Colombia and Venezuela, and one from southeastern Brazil. The only known host plant belongs to the Sapotaceae.

5.2.5. The robusta Group

The following character states appear to be synapomorphies of the robusta group: 1) scutum with dark brown band on posterior margin in most species (Fig. 1C) (apomorphic, but this band is sometimes faint, and a similar one occurs in a few species in other species groups, e.g., A. cryptostrepha); 2) eversible membrane with relatively few, stout dorsobasal teeth (Fig. 13D, 15C-D) (apomorphic, but numerous teeth are present in A. speciosa (Fig. 15E) and two related new species, possibly via reversal); and 3) the two species for which eggs have been examined (A. nigrifascia and pittieri) have a long lobe on the micropyle end (Fig. 3C) (apomorphic, but this occurs, apparently by convergence, within the spatulata group and in other tephritid genera). Many of the species in this group (fenestrata, lambda, pittieri, speciosa, n. sp. near lambda, and two new species near speciosa) have a unique apomorphic wing character: There is an extension from the middle of the S-band that reaches the posterior wing margin in the middle of cell cu₁ (Fig. 5H). This character further supports the inclusion of speciosa and the two new related species in the robusta group, despite their different eversible membranes. There are at least five undescribed species belonging to the robusta group (Norrbom, pers. obs.). Tigrero (1998) included A. concava and A. montei in his fenestrata group (A. fenestrata is here included in the robusta group). We place A. montei in the spatulata group. A. concava, which has a similar wing pattern to the robusta group (generally dark brown and with a distinct cleft in the base of the S-band) and undivided proctiger sclerotization, and the little studied species A. connexa, hamadryas and longicauda might possibly be related to the robusta group or to the benjamini group, all of which have the basal cleft in the S-band. They differ from the robusta group in having long terminalia and relatively long lateral surstyli.

The robusta group is widespread, including species from Mexico to Brazil, with two species in the Antilles and Florida. The known host plants (for only 2 spp.) are Moraceae and Sapotaceae.

5.2.6. The schausi Group

The following unique apomorphies indicate that this group is monophyletic: 1) glans with minute spines (Fig. 11F); 2) facial carina weak; and 3) lateral surstylus acute, lateral margin usually slightly concave (Fig. 9K). All of the species appear to be sexually dimorphic, with the male face (Fig. 4D), abdomen (Fig. 1D), and/or wing having unusual markings or pattern, although not all of the species have the same dimorphic structures, and whether this can be considered a single homologous character is uncertain. The male of A. bellicauda has only some diffuse yellow markings on its wing (Fig. 5J), but the recently discovered female has a more normal pattern in which all of the typical Anastrepha bands can be recognized (Fig. 5I). This group was revised by Norrbom & Kim (1988a). The four included species are known from Costa Rica, Panama, Venezuela, Peru and Bolivia. A. bellicauda is the only species to have been reared, but the identity of the host is uncertain, although it is probably a species of Sapotaceae (K. P. Katiyar, pers. comm.).

5.2.7. The punctata Group

Two unique apomorphies support this group as monophyletic: 1) scutum with pair of posterior brown spots (Fig. 1E) (variable within an undescribed species near A. luederwaldti; brown markings in this area occur rarely in other species groups, e.g., in A. dentata and pallens, but differ in shape); and 2) lateral surstylus obliquely oriented, strongly curved and blunt apically (Fig. 9I-J). This small group is known only from southern Brazil, Paraguay and Argentina. The only known host plant belongs to the Myrtaceae.

5.2.8. The leptozona Group

Two synapomorphies indicate that this group is monophyletic: 1) lateral surstylus short and somewhat boot-shaped, with laterally projecting apical lobe (Fig. 9L) (apomorphic); 2) vein M extremely strongly curved (Fig. 5L) (occurs convergently in a few species in other species groups, e.g., A. montei). This small group is widely distributed in the mainland Neotropics. Two of the four species have known hosts, and both attack Sapotaceae, although A. leptozona breeds in a range of other plant families as well.

5.2.9. The mucronota Group

This is perhaps a paraphyletic group; it includes species with the male proctiger not strongly creased (plesiomorphic), but not fitting in any other species group above. But it might be monophyletic, and at least seems to include some groups of closely related species. The wing bands are usually well separated along vein R₄₊₅ (Fig. 6A-B) (perhaps a synapomorphy, but this character is very variable across the genus and also occurs in the leptozona, punctata and schausi groups), and vein R₂₊₃ tends to be sinuous (Fig. 6B) (apomorphic, but varies, often intraspecifically). Anastrepha minuta has an unusual character of the eversible membrane that may have behavioral significance: On the ventral side, opposite the hook-like dorsobasal teeth, it has a large clump of fine, hair-like projections (Fig. 13I). The parallela and integra groups of Tigrero (1998) fit here; he included A. mucronota in the former group.

The mucronota group is widespread, from Mexico to Argentina, with four species known from the Antilles. Five of the species, so far as known, breed only in Bombacaceae. Six others have reported hosts belonging to the Sapotaceae (for 4 spp.), Sterculiaceae (for 2 spp.), Annonaceae (2 spp.), Passifloraceae and Rutaceae.

5.2.10. The grandis Group

This group is rather weakly supported by the following apomorphies and may not be monophyletic: 1) wing with complete marginal band, without hyaline mark at apex of vein R₁ (Fig. 6C-D) (absent in 2 species; occurs convergently in bezzii, cordata, and within daciformis group); 2) distal arm of V-band absent (this also occurs within various other species groups); and 3) body color relatively dark (Fig. 1H) (this also occurs in several other species groups). Anastrepha bivittata and fumipennis lack the first apomorphy (they have a hyaline marginal spot in cell r₁ (Fig. 6E)), but are included in this group because of other apomorphies shared with A. atrigona, including: 1) cell bm infuscated (Fig. 6D-E) (occurs rarely in other Anastrepha, e.g., flavipennis); 2) crossvein R-M relatively close to DM-Cu (Fig. 6D-E); and 3) aculeus tip broadly rounded (Fig. 16J) (Norrbom 1991). The hypothesis that the presence of the hyaline spot in r₁ in A. bivittata and fumipennis is due to reversal (i.e., the grandis group is monophyletic) is equally parsimonious with the hypothesis that the complete marginal band (absence of a hyaline spot in r₁) in A. atrigona is due to homoplasy (i.e., the grandis group is not monophyletic), and further testing of the monophyly of this group is needed. A. aberrans and a related undescribed species from Costa Rica lack the basal half of the S-band unlike the other species of the grandis group (apomorphic; occurs convergently within the daciformis group), and resemble A. castilloi in the pattern of the dorsobasal teeth of the eversible membrane. Steyskal (1977b) included A. bezzii in this group based on similarity in wing pattern, but Norrbom & Kim (1988b; also see Norrbom 1991) transferred it to the mucronota group based on genitalic and other characters. Except for the new species near A. aberrans, the grandis group is restricted to South America. Host plants are known only for A. grandis, which breeds in Cucurbitaceae.

5.2.11. The doryphoros Group

The following synapomorphies indicate that this group is monophyletic: 1) scutum with paired nonmicrotrichose submedial stripe (in most Anastrepha species, the scutum is almost entirely microtrichose or is mostly without microtrichia, although shorter, broader bare stripes occur in the striata group and some species of the schausi group); 2) wing with apical part of S-band broadly fused to V-band (Fig. 6F), completely fused in cell r₂₊₃, and in cell r₄₊₅ separated at most by a hyaline area that does not extend to vein R₄₊₅, in cell m hyaline spot between arms of V-band not extended beyond vein M; 3) eversible membrane with dorsobasal teeth short and arranged in elongate pattern (Fig. 14A) (A. bezzii (Fig. 15F) has a somewhat similar elongate pattern, but the teeth are larger); and 4) aculeus tip tapered, then parallel-sided, then tapered (Fig. 16L). This group includes a new species from Costa Rica in addition to A. doryphoros and A. freidbergi, which were revised by Norrbom (1993). Anastrepha conjuncta, known only from the male holotype, might be related to this group. As in the species of the doryphoros group, the medial scutal pale stripe is very narrow and not expanded posteriorly, and the genitalia are very long. The female of conjuncta is unknown, but the phallus of the holotype is over 15 mm. long. This species lacks the other non-genitalic apomorphies of the doryphoros group, however. The scutum is entirely microtrichose, and, although the S-band is broadly fused to both the C-band and V-band, the wing pattern is typical for Anastrepha. Discovery of the female may help clarify the relationships of A. conjuncta. The doryphoros group occurs in Costa Rica, Panama, Colombia, Peru and Bolivia. The host plants for the group are unknown.

5.2.12. The spatulata Group

The following two character states may be synapomorphies of this group: 1) aculeus tip short and broad (except in montei and haywardi in which the tip is extremely slender), with numerous fine serrations extended beyond the base (less extensively in haywardi) (Fig. 16M); and 2) in the two species whose karyotype has been studied (montei and pickeli), the diploid number (2n) is eight, whereas in most other Anastrepha species studied and in Tephritidae in general the diploid number is 12 (Solferini & Morgante 1987, Morgante et al. 1996). The status of the spatulata group needs further testing, as the former character state does not occur in all of the species (further evolution of this character in A. montei and haywardi must be postulated if the state described above is a synapomorphy for the entire species group), and the latter is unknown for most of the species, although it is probably a synapomophy for at least the Manihot-infesting species. A. nascimentoi, rheediae, and tecta, which have broad and finely serrate aculeus tips, may also belong in the spatulata group, although their tips are slightly longer and somewhat intermediate in shape between species of the spatulata and pseudoparallela groups. Tigrero (1998) placed A. montei (probably based on a misidentified female of the A. pickeli complex according to his figures of the aculeus and wing, especially the shape of vein M) in his fenestrata group, and A. manihoti in his manihoti subgroup of the chiclayae group. These three taxa, which breed in Manihot (Euphorbiaceae), and A. haywardi, whose host is unknown, appear instead to form a closely related group of species. A. manihoti and pickeli have elongate lobes on their eggs, but such lobes do not occur on the eggs of A. alveatoides and haywardi, so this is not a synapomorphy for the entire spatulata group.

The spatulata group occurs from Texas to Argentina, although it appears to be absent from most of Amazonia. One species occurs in the Antilles and Florida. Host plants are known for seven species, two of which breed in Olacaceae, the other five in Euphorbiaceae (one species also in Bombacaceae).

5.2.13. The ramosa Group

A single unique apomorphy indicates that this small species group is monophyletic: Aculeus tip short, broad, and triangular, with very large serrations extended to base (Fig. 17F). This group includes only two species, from Central America and Panama, whose hosts are unknown.

5.2.14. The pseudoparallela Group

This group is recognized mainly on the basis of aculeus tip shape, but its status needs further testing, as the single putative synapomorphy here proposed is not present in all species. All but three species have the aculeus tip long, moderately broad, and mostly or entirely finely serrate (Fig. 16N-O) (except nonserrate in velezi, and much shorter in passiflorae and especially xanthochaeta), which is possibly a synapomorphy, although similar aculeus shapes occur in the leptozona and serpentina groups, so it may alternatively be a synapomorphy at a higher level. This group includes the species placed by Norrbom & Kim (1988b) in the pseudoparallela and chiclayae groups, which Norrbom (1997) considered to intergrade. A. velezi is included in this group because it has what Norrbom (1997) hypothesized to be a unique synapomorphy of the pallidipennis complex, which also includes A. amnis, curitis, pallida, and pallidipennis: The spiracle of the female oviscape is located very close to its base (Fig. 14D). Some species of the pallidipennis complex resemble species of the benjamini group in facial carina shape (Fig. 4F) and were placed in that group by Steyskal (1977b), but because this character occurs in a few other species that otherwise appear to belong to other species groups, Norrbom (1997) regarded it as more likely to be homoplastic than the oviscape character, and transferred the species of the pallidipennis complex to this species group. Based on host plant data, we have tentatively included A. mburucuyae, which was described from only males, as well as A. xanthochaeta and passiflorae, in the pseudoparallela group. These three species and the other 11 species (of 20 spp.) of the group for which host data are known breed in Passifloraceae (two species have reported hosts in other families in addition). Tigrero (1998) proposed the palae group for A. townsendi, which we include here, placed A. pseudoparallela in a subgroup of the distincta group (here considered part of the fraterculus group), and continued to recognize the chiclayae group, in which he also included A. manihoti. The pseudoparallela group is widespread, from Texas to Argentina, although only two species extend north of Panama and one into the Antilles.

5.2.15. The serpentina Group

This group is supported by the following synapomorphies: 1) abdomen bicolored, partially dark brown (Fig. 1J) (abdominal markings also occur in some species of the cryptostrepha, schausi, punctata, grandis and daciformis groups, but differences in the patterns and/or other character state distributions (e.g., the synapomorphies of the daciformis group) suggest that this is homoplasy, except perhaps in A. ornata (see striata group)); 2) aculeus tip long, moderately broad, evenly tapered, and partially finely serrate (Fig. 16P, 17G) (probably apomorphic, but this is an uncertain synapomorphy at this level because similar aculeus tip shapes occur in the leptozona and pseudoparallela groups); and 3) wing pattern relatively dark brown (Fig. 6I) (occurs in occasional other Anastrepha species, e.g., within robusta group). Steyskal (1977b) also included A. ornata, fenestrata and phaeoptera in this group. We include the former in the striata group, although it strongly resembles A. serpentina in wing pattern. Norrbom (1985) transferred A. fenestrata and phaeoptera to the robusta group, although the unique male holotype of the latter species should be restudied to confirm this action.

The serpentina group occurs mainly between Guatemala and Brazil, although A. serpentina extends to Texas and Argentina, and A. ocresia occurs in the Antilles and Florida. Host plants are known for five of the seven species, but vary considerably. Three species breed in Apocynaceae, and two in Sapotaceae (including A. serpentina, which has hosts in both families). The only native host known for A. ocresia belongs to the Myrtaceae, but it has been reared from an exotic species of Sapotaceae as well.

5.2.16. The striata group

The following apomorphies support this group as monophyletic: 1) aculeus tip extremely broad and gradually tapered to blunt point (Fig. 17H-I) (unique apomorphy); and 2) scutum mostly microtrichose, with sublateral bare stripes (see discussion of this character under doryphoros group). The striata group includes A. ornata, from Colombia and Ecuador, and the disjunct pair, A. striata, from Mexico to Bolivia and northern Brazil, and A. bistrigata, from southeastern Brazil. The latter two species, which have very similar male genitalia (Fig. 10K-L) and body markings (Fig. 1I), are probably sister taxa. A. ornata is placed in this group based on the two hypothesized synapomorphies listed above, but it also resembles species of the serpentina group in abdominal color and wing pattern (Fig. 6J). It may possibly belong in the serpentina group (Steyskal 1977b, Tigrero 1998), or these two species groups may be related, although there is homoplasy in at least some of the characters discussed above in any hypothesis of relationships among the species of the serpentina and striata groups. The single known native host plant of A. ornata, and most of those of A. striata and bistrigata, belong to the family Myrtaceae.

5.2.17. The fraterculus Group

The following possible synapomorphies suggest that this group is monophyletic: 1) mediotergite and/or subscutellum with lateral dark brown stripe or spot and body otherwise without dark markings, except sometimes on scuto-scutellar suture (Fig. 2E-F) (apomorphic, but absent in two species and variable in a few others); 2) aculeus tip usually partially serrate and often with constriction just basal to serrate part (Fig. 12D, 16Q-R) (neither attribute is unique to this group, but this aculeus shape may be a synapomorphy, depending upon what is the sister group of the fraterculus group); and 3) lateral surstylus usually parallel sided or slightly tapered and truncate apically (Fig. 10M-O) (except in schultzi; apomorphic, but possibly not unique to the group). Norrbom (1985) attempted to divide the species placed here into two groups (distincta and fraterculus groups), but they have been found to intergrade. All of the included species except A. barbiellinii and zuelaniae at least sometimes have the lateral brown markings on the mediotergite or subscutellum. This includes A. antunesi, of which several specimens from Venezuela with lateral markings on the mediotergite have been observed (Norrbom, pers. obs.). Similar markings occur in various species of Anastrepha in which other areas of the body are dark (e.g., in the daciformis, striata, and serpentina groups), but such markings are otherwise rare in predominantly yellow species. An exception is A. rheediae, which might belong in the spatulata or pseudoparallela groups based on aculeus tip shape. A. zuelaniae is included here on the basis of its surstylus shape and aculeus tip shape. The egg of A. barbiellinii is very similar to that of A. obliqua (Fig. 3D), having a short lobe on the micropyle end, and this species is tentatively included in the group on that basis. The fraterculus group is widespread, from Texas to Argentina, with two species occurring in the Antilles. The host plants are very diverse, and seven of the species are reported to have native hosts in three or more families.

5.2.18. Unplaced Species

Many of the 32 species which have not been placed in a species group are poorly studied and/or are known from only one sex. Tigrero (1998) recognized the nigripalpis group for A. nigripalpis, and the bondari group for A. buscki, but without further explanation. As noted above, A. conjuncta may be related to the doryphoros group, and A. concava, connexa, hamadryas and longicauda share certain characters with the robusta and benjamini groups and may be related to one or both of them or form another species group. Species such as A. acris, buscki, and palae, which have distinctly creased proctigers in the male, belong to the large clade also including the pseudoparallela + spatulata + ramosa + doryphoros + grandis + serpentina + striata + fraterculus groups. Tigrero (1998) placed A. tecta and rheediae in his rheediae group, presumably based on their somewhat similar aculeus tips. These species are intermediate in this character between the pseudoparallela and spatulata groups, and we prefer to leave them unplaced at this time.

6. HOST PLANT RELATIONSHIPS

Published host plant records for Anastrepha and Toxotrypana have been compiled in a database (Norrbom, in prep.). Fields were included for the reliability of the record (based on Norrbom’s subjective evaluation) and the origin of the plant (exotic or native, based on whether or not its original distribution coincided with that of the fly species). Only native hosts and reliable records were used to produce the host plant summary in Table 4, in which the host family or families are listed for each fly species. It was augmented by data from theses or unpublished sources contributed by Zucchi for 15 species from Brazil. Several trends in host relationships can be observed based on this data.

Most species of Hexachaeta and all species of Toxotrypana for which host plants are known attack latex-bearing plants. The hosts for the nine Hexachaeta species that have been reared belong to the Moraceae, except one species was reported from a Bignoniaceae species and another has been reported from fruit of one Verbenaceae and one Moraceae species (Silva et al. 1968, Foote et al. 1993, Hernández-Ortiz, unpub. data). The Bignoniaceae and Verbenaceae are not latex bearing. The known hosts of Toxotrypana species belong to the Caricaceae, Asclepiadaceae, and possibly Apocynaceae; at least two species attack hosts in both of the former families. Toxotrypana larvae feed on the developing seeds and ovaries within the chamber inside the fruit (the larvae are killed if they enter the flesh of papaya fruit before it is ripe (Knab & Yothers 1914)), whereas those of at least one undescribed species of Hexachaeta feed inside the seeds within the fruit of their Sorocea host plant (Moraceae) (Norrbom, pers. obs.).

Many Anastrepha species, including the majority of those in the primitive clades, also breed in fruits of latex-bearing plants, especially Sapotaceae. Certain species groups are associated mainly with other plant families, such as the spatulata group on Euphorbiaceae and Olacaceae, the pseudoparallela group on Passifloraceae, the grandis group (or at least the one species of the group whose hosts are known) on Cucurbitaceae, the striata group on Myrtaceae, and some species of the mucronota group on Bombacaceae. In most published host plant records for Anastrepha, the part of the fruit attacked has not been mentioned. In future rearing studies, this would be useful for researchers to observe and report. At least in the case of A. sagittata (dentata group) and A. katiyari and pallens (daciformis group), the larva feeds exclusively in the seed (McPhail & Berry 1936, Baker et al. 1944, K. P. Katiyar, pers. comm.). Larvae of A. pseudoparallela (pseudoparallela group), and A. montei and pickeli (spatulata group) feed on developing seeds and associated tissues (Morgante et al. 1996, Stefani & Morgante 1996) similar to species of Toxotrypana. Larvae of A. cordata (cryptostrepha group) feed mainly on the seeds and later on the pulp of the fruit, apparently because immature fruits have large quantities of latex (Hernández-Ortiz & Pérez-Alonso 1993). Larvae of A. crebra (mucronota group) have been found feeding in the pulp and seeds at same time (Hernández-Ortiz & Pérez-Alonso 1993), and those of A. steyskali (leptozona group), and A. anomala and serpentina (serpentina group) reportedly can feed on both seeds and the fleshy mesocarp (Stone 1942, Korytkowski 1974). Conversely, in most of the generalist species, such as A. obliqua, fraterculus, striata and distincta, feeding is primarily or exclusively on the mesocarp (Aluja 1994). The effects on behavior that may result from these different larval feeding modes may be worthy of study. For example, females of Toxotrypana curvicauda and Anastrepha cordata, which are seed feeders, do not need to forage for protein in order to produce eggs, unlike most of the pulp feeding, pest species of Anastrepha (Landolt 1985, Aluja 1994).

It is noteworthy that most of the generalist, pest species of Anastrepha belong to the fraterculus species group. Except for single species in the serpentina, striata, leptozona and pseudoparallela groups, feeding on more than a few related hosts or 1-2 families by a single Anastrepha species is rare outside of the fraterculus group. The reasons for the preponderance of pest species in this one group may be varied, but there may be some connection to the larval feeding mode. Perhaps the specialization of the species of the fraterculus group towards feeding on a variety of pulpy, mature fruits, rather than on a narrower range of plants with particular toxic chemicals, has preadapted them for attacking cultivated fruits.

7. CONCLUSIONS AND FUTURE RESEARCH NEEDS

One way to assess our current understanding of the phylogeny of Anastrepha and Toxotrypana is to compare with other tephritid genera. In that sense, Anastrepha and Toxotrypana are better understood than most Tephritidae, but they have not been as well studied as Rhagoletis, the most intensely studied tephritid genus.

The monophyly of Anastrepha + Toxotrypana is well supported by morphological characters, as is that of Toxotrypana. Whether Anastrepha is monophyletic remains uncertain. Based mainly on morphological studies of the adult stage, 18 species groups, including 164 valid species, have been recognized within Anastrepha (another 32 species are unplaced). There is strong to fairly good character evidence (hypothesized synapomorphies), in roughly decreasing order, for the A. daciformis, schausi, dentata, punctata, leptozona, doryphoros, ramosa, robusta, fraterculus, striata, serpentina, and spatulata groups. The status of the A. cryptostrepha, grandis, pseudoparallela, benjamini, and mucronota groups is less certain, although each contains some species that are clearly closely related. Relationships among the species groups and within most of them are poorly resolved.

Among the most important questions remaining in the phylogeny of these two genera are: whether Anastrepha is monophyletic or if some species are more closely related to Toxotrypana; what are the relationships among the species groups of Anastrepha; and for most of the groups, what are the relationships among the species. To answer these questions, much additional morphological study is needed to treat the many undescribed species remaining in both genera, and for many of the known species, to fully describe additional characters, such as the male genitalia. Sources of additional character data, such as larvae, eggs, and molecular characters need to be further explored. Determination of the natural host plants of the as yet biologically unknown species will be crucial to obtain the immature stages of these species and to provide other useful data, such as the part of the plant attacked (particularly, what tissue(s) within the fruit). Based on all these additional data and currently known information, a rigorous cladistic analysis is needed for the entire group.

8. ACKNOWLEDGMENTS

We are grateful to the many scientists and institutions, too numerous to list here, who have loaned or sent us specimens for study. We thank Sonja Scheffer, Marc De Meyer, and Norman Woodley for reviewing the manuscript. We acknowledge the Campaña Nacional Contra las Moscas de la Fruta (Mexico), International Organization for Biological Control of Noxious Animals and Plants (IOBC), and Instituto de Ecología, A.C. (Mexico) for their financial support of the Symposium, and USDA-OICD-RSED for funding Norrbom’s travel.

9. LITERATURE CITED

Aldrich, J. M. 1925. New Diptera or two-winged flies in the United States National Museum. Proc. U.S. Natl. Mus. 66 (18): 36 p. [= No. 2555].

Aluja, M. 1994. Bionomics and management of Anastrepha. Annu. Rev. Entomol. 39: 155-178.

Baker, A. C., W. E. Stone, C. C. Plummer & M. McPhail. 1944. A review of studies on the Mexican fruitfly and related Mexican species. U.S. Dep. Agric. Misc. Publ. 531: 155 p.

Canal D., N. A. 1997. Levantamento, flutuação populacional e análise faunística das espécies de moscas-das-frutas (Dip., Tephritidae) en quatro municípios do Norte do Estado de Minas Gerais. Dissertation, Universidade de São Paulo, Piracicaba, 113 p.

Carroll, L. E. & R. A. Wharton. 1989. Morphology of the immature stages of Anastrepha ludens (Diptera: Tephritidae). Ann. Entomol. Soc. Am. 82: 201-214.

Emmart, E. W. 1933. The eggs of four species of fruit flies of the genus Anastrepha. Proc. Entomol. Soc. Wash. 35: 184-191.

Foote, R. H. 1967. Family Tephritidae (Trypetidae, Trupaneidae). In A Catalogue of the Diptera of the Americas South of the United States (P. E. Vanzolini & N. Papavero, eds.), fasicle 57: 91 p. Departamento de Zoologia, Secretaria da Agricultura, São Paulo.

Foote, R. H., F. L. Blanc & A. L. Norrbom. 1993. Handbook of the Fruit Flies (Diptera: Tephritidae) of America North of Mexico. Comstock Publishing Associates, Cornell University Press, Ithaca. xii + 571 p.

Freidberg, A. 1985. The genus Craspedoxantha Bezzi (Diptera: Tephritidae: Terelliinae). Ann. Natal Mus. 27: 183-206.

Frías L., D., H. Martinez & A. Alviña. 1993. Descripcion morfologica de los estados inmaduros de Rhagoletis tomatis Foote (Diptera: Tephritidae). Acta Entomol. Chil. 18: 31-40.

Han, H. Y. & B. A. McPheron. 1997. Molecular phylogenetic study of Tephritidae (Insecta: Diptera) using partial sequences of the mitochondrial 16S ribosomal DNA. Mol. Phylogenet. Evol. 7: 17-32.

Hancock, D. L. 1986. Classification of the Trypetinae (Diptera: Tephritidae), with a discussion of the Afrotropical fauna. J. Entomol. Soc. South. Afr. 49: 275-305.

Hardy, D. E. 1955. A reclassification of the Dacini (Tephritidae-Diptera). Ann. Entomol. Soc. Am. 48: 425-437.

Headrick, D. H. & R. D. Goeden. 1990. Description of the immature stages of Paracantha gentilis (Diptera: Tephritidae). Ann. Entomol. Soc. Am. 83: 220-229.

Headrick, D. H. & R. D. Goeden. 1993. Life history and description of immature stages of Aciurina thoracica (Diptera: Tephritidae) on Baccharis sarothroides in southern California. Ann. Entomol. Soc. Am. 86: 68-79.

Heppner, J. B. 1986. Larvae of fruit flies. III. Toxotrypana curvicauda (Papaya fruit fly) (Diptera: Tephritidae). Fla. Dep. Agric. Consum. Serv. Div. Plant Ind. Entomol. Circ. 282: 2 p.

Hernández-Ortiz, V. & R. Pérez-Alonso. 1993. The natural host plants of Anastrepha (Diptera: Tephritidae) in a tropical rain forest of Mexico. Fla. Entomol. 76: 447-460.

Kitto, G. B. 1983. An immunological approach to the phylogeny of the Tephritidae. In Fruit Flies of Economic Importance. Proceedings of the CEC/IOBC International Symposium, Athens, Greece, November 16-19, 1982 (R. Cavalloro, ed.), p. 203-211. A. A. Balkema, Rotterdam. xii + 642 p.

Knab, F. & W. W. Yothers. 1914. Papaya fruit fly. J. Agric. Res. 2: 447-453.

Korneyev, V. A. 1994. Reclassification of the Palaearctic Tephritidae (Diptera). Communication 2. Vestn. Zool. 1994 (1): 3-17.

Korytkowski G., C. A. 1974. Una nueva especie del genero Anastrepha Schiner (Diptera: Tephritidae). Rev. Peru. Entomol. 17: 1-3.

Korytkowski G., C. A. & D. Ojeda P. 1968. Especies del genero Anastrepha Schiner 1868 en el nor-oeste peruano. Rev. Peru. Entomol. 11: 32-70.

Landolt, P. J. 1985. Behavior of the papaya fruit fly Toxotrypana curvicauda Gerstaecker (Diptera: Tephritidae), in relation to its host plant, Carica papaya L. Folia Entomol. Mex. (1984) 61: 215-224.

Lima, A. M. da Costa. 1953a. Moscas de frutas do U.S. National Museum (Smithsonian Institution) (II) (Diptera: Trypetidae). An. Acad. Bras. Cienc. 25: 153-155.

Lima, A. M. da Costa. 1953b. Moscas de frutas do U.S. National Museum (Smithsonian Institution) (III) (Diptera, Trypetidae). An. Acad. Bras. Cienc. 25: 557-566.

Lima, A. M. da Costa. 1954. Moscas de frutas do U.S. National Museum (Smithsonian Institution) (IV) (Diptera, Trypetidae). An. Acad. Bras. Cienc. 26: 277-282.

Loew, H. 1873. Monographs of the Diptera of North America. Part III. Smithson. Misc. Collect. 11 (3 [= pub. 256]): vii + 351 + XIII p.

McPhail, M. & N. O. Berry. 1936. Observations on Anastrepha pallens (Coq.) reared from wild fruits in the lower Rio Grande Valley of Texas during the spring of 1932. J. Econ. Entomol. 29: 405-411.

Morgante, J. S., D. Selivon, V. N. Solferini & A. S. do Nascimento. 1996. Genetic and morphological differentiation in the specialist species Anastrepha pickeli and A. montei. In Fruit Fly Pests: A World Assessment of their Biology and Management (B. A. McPheron & G. J. Steck, eds.), p. 259-276. St. Lucie Press, Delray Beach. 586 p.

Murillo, T. & L. F. Jirón. 1994. Egg morphology of Anastrepha obliqua and some comparative aspects with eggs of Anastrepha fraterculus (Diptera: Tephritidae). Fla. Entomol. 77: 342-348.

Norrbom, A. L. 1985. Phylogenetic analysis and taxonomy of the cryptostrepha, daciformis, robusta, and schausi species groups of Anastrepha Schiner (Diptera: Tephritidae). Dissertation, Pennsylvania State University, University Park. xvi + 354 p.

Norrbom, A. L. 1991. The species of Anastrepha (Diptera: Tephritidae) with a grandis-type wing pattern. Proc. Entomol. Soc. Wash. 93: 101-124.

Norrbom, A. L. 1993. Two new species of Anastrepha (Diptera: Tephritidae) with atypical wing patterns. Proc. Entomol. Soc. Wash. 95: 52-58.

Norrbom, A. L. 1997. Revision of the Anastrepha benjamini species group and the A. pallidipennis complex (Diptera: Tephritidae). Insecta Mundi 11: 141-157.

Norrbom, A. L. 1998. A revision of the Anastrepha daciformis species group (Diptera: Tephritidae). Proc. Entomol. Soc. Wash. 100: 160-192.

Norrbom, A. L., L. E. Carroll & A. Freidberg. 1999a. Status of knowledge. In Fruit Fly Expert Identification System and Systematic Information Database (F.C. Thompson, ed.), pp. 9-47. Myia (1998) 9: 524 pp.

Norrbom, A. L., L. E. Carroll, F. C. Thompson, I. M. White & A. Freidberg. 1999b. Systematic database of names. In Fruit Fly Expert Identification System and Systematic Information Database (F.C. Thompson, ed.), pp. 65-251. Myia (1998) 9: 524 pp.

Norrbom, A. L. & R. H. Foote. 1989. The taxonomy and biogeography of the genus Anastrepha (Diptera: Tephritidae). In Fruit Flies, Their Biology, Natural Enemies, and Control (A. S. Robinson & G. Hooper, eds.), p. 15-26. In World Crop Pests (W. Helle, ed.), vol. 3A. Elsevier Science Publishers, Amsterdam. 372 p.

Norrbom, A. L. & K. C. Kim. 1988a. 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: 164-173.

Norrbom, A. L. & K. C. Kim. 1988b. A list of the reported host plants of the species of Anastrepha Schiner (Diptera: Tephritidae). U. S. Dep. Agric. Animal Plant Health Insp. Serv. no. 81-52, 114 p.

Phillips, V. T. 1946. The biology and identification of trypetid larvae (Diptera: Trypetidae). Mem. Am. Entomol. Soc. 12: 161 p.

Sarma, R., G. B. Kitto, S. H. Berlocher & G. L. Bush. 1987. Biochemical and immunological studies on an alpha-glycerophosphate dehydrogenase from the tephritid fly, Anastrepha suspensa. Arch. Insect Biochem. Physiol. 4: 271-286.

Seín, F., Jr. 1933. Anastrepha (Trypetydae [sic], Diptera) fruit flies in Puerto Rico. J. Dep. Agric. P. Rico 17: 183-196.

Selivon, D., J. S. Morgante & A. L. P. Perondini. 1997. Egg size, yolk mass extrusion and hatching behavior in two cryptic species of Anastrepha fraterculus (Wiedemann) (Diptera, Tephritidae). Braz. J. Genet. 91: 471-478.

Selivon, D. & A. L. P. Perondini. 1999. Description of Anastrepha sororcula and A. serpentina (Diptera, Tephritidae) eggs. Fla. Entomol. 82: 347-353.

Shaw, J. G. 1962. Species of the spatulata group of Anastrepha (Diptera: Tephritidae). J. Kans. Entomol. Soc. 35: 408-414.

Silva, A. G. d'Araujo e, C. R. Goncalves, D. M. Galvao, A. J. L. Goncalves, J. Gomes, M. do Nascimento Silva & L. de Simoni. 1968. Quarto catalogo dos insetos que vivem nas plantas do Brasil. Seus parasitos e predadores. Parte II - 1.o Tomo. Insetos, hospedeiros, e inimigos naturais. Ministerio da Agricultura, Departamento de Defesa e Inspecao Agropecuaria, Servico de Defesa Sanitaria Vegetal, Laboratorio Central de Patologia Vegetal, Rio de Janeiro. vi-xxiv + 1-622 p.

Snow, W. A. 1895. On Toxotrypana of Gerstaecker. Kans. Univ. Q. 4: 117-119.

Solferini, V. N. & J. S. Morgante. 1987. Karyotype study of eight species of Anastrepha (Diptera: Tephritidae). Caryologia 40: 229-241.

Steck, G. J. 1991. Biochemical systematics and population genetic structure of Anastrepha fraterculus and related species (Diptera: Tephritidae). Ann. Entomol. Soc. Am. 84: 10-28.

Steck, G. J., L. E. Carroll, H. Celedonio H. & J. C. Guillen A. 1990. Methods for identification of Anastrepha larvae (Diptera: Tephritidae), and key to 13 species. Proc. Entomol. Soc. Wash. 92: 333-346.

Steck, G. J. & A. Malavasi. 1988. Description of immature stages of Anastrepha bistrigata (Diptera: Tephritidae). Ann. Entomol. Soc. Am. 81: 1004-1006.

Steck, G. J. & R. A. Wharton. 1988. Description of immature stages of Anastrepha interrupta, A. limae, and A. grandis (Diptera: Tephritidae). Ann. Entomol. Soc. Am. 81: 994-1003.

Stefani, R. N. & J. S. Morgante. 1996. Genetic variability in Anastrepha pseudoparallela: A specialist species. In Fruit Fly Pests: A World Assessment of their Biology and Management (B. A. McPheron & G. J. Steck, eds.), p. 259-276. St. Lucie Press, Delray Beach. 586 p.

Steyskal, G. C. 1977a. Two new neotropical fruitflies of the genus Anastrepha, with notes on generic synonymy (Diptera, Tephritidae). Proc. Entomol. Soc. Wash. 79: 75-81.

Steyskal, G. C. 1977b. Pictorial Key to Species of the Genus Anastrepha (Diptera: Tephritidae). Entomological Society of Washington, Washington, D.C. 35 p.

Stone, A. 1939a. A new genus of Trypetidae near Anastrepha (Diptera). J. Wash. Acad. Sci. 29: 340-350.

Stone, A. 1939b. A revision of the genus Pseudodacus Hendel (Dipt. Trypetidae). Rev. Entomol. (Rio J.) 10: 282-289.

Stone, A. 1942. The fruitflies of the genus Anastrepha. U.S. Dep. Agric. Misc. Publ. 439: 112 p.

Tauber, M. J. & C. A. Tauber. 1967. Reproductive behavior and biology of the gall-former Aciurina ferruginea (Doane) (Diptera: Tephritidae). Can. J. Zool. 45: 907-913.

Tigrero, J. 1992. Descripción de dos nuevas especies de Tephritidae: Toxotrypaninae, presentes en Ecuador. Rev. Rumipamba 9: 102-112.

Tigrero, J. 1998. Revisión de Especies de Moscas de la Fruta Presentes en el Ecuador. Published by the author, Sangolquí, Ecuador, 55 pp.

White, I. M. & M. M. Elson-Harris. 1992. Fruit Flies of Economic Significance: Their Identification and Bionomics. International Institute of Entomology, London. xii + 601 p.

White, I. M. & K. Marquardt. 1989. A revision of the genus Chaetorellia Hendel (Diptera: Tephritidae) including a new species associated with spotted knapweed, Centaurea maculosa Lam. (Asteraceae). Bull. Entomol. Res. 79: 453-487.

Zucchi, R. A. 1977. Taxonomia das especies brasileiras de Anastrepha Schiner, 1868 do compexo fraterculus (Diptera, Tephritidae). Thesis, Universidade de São Paulo, Piracicaba. 63 p.

Table 1. Dimensions (in mm) and shape of eggs of Anastrepha and Toxotrypana species.

Note: For species with lobes, length indicates that of the main part of the egg, not including the lobe. Except for A. ludens, data for species from Norrbom (1985) or not cited from other publications are based on eggs dissected by Norrbom from abdomens of females.

* Includes illustration of egg.

** Total egg length; length of lobe not stated.

Table 2. Morphological characters useful for phylogenetic analysis in Toxotrypana and Anastrepha. The plesiomorphic characater state is coded 0.

1. Body color - 0) predominantly yellow or orange; 1) mostly dark orange to dark brown.

2. Facial carina - 0) not strongly produced medially or dorsally, concave or straight in lateral view; 1) strongly produced medially, convex in lateral view; 2) strongly produced dorsally; 3) weak, indistinct.

3. Ocellar seta - 0) well developed; 1) small and weak or absent.

4. Frontal, orbital, postpronotal, acrostichal, dorsocentral, presutural supra-alar, anepisternal, katepisternal, and basal scutellar setae - 0) well developed; 1) very small and weak or absent. Acrostichal setae are variable in A. cordata and absent in an undescribed species from Costa Rica.

5. Scutum with a medial longitudinal depression - 0) no; 1) yes.

6. Scutum - 0) without brown markings, or if present, in different pattern; 1) without brown markings except band on posterior margin; 2) with 2 pairs of dark brown stripes and band on posterior margin; 3) without brown markings except paired, circular spot near posterolateral corner. Additional types of scutal color patterns occur within Hexachaeta, and the daciformis, dentata, grandis, serpentina, and striata species groups. Some species in the daciformis and dentata groups have irregular, often acute spots in the area of the circular spots of the punctata group.

7. Scutal microtrichia - 0) mostly or entirely microtrichose; 1) mostly or entirely bare of microtrichia; 2) microtrichose except dorsocentral bare stripe. The coding of this character is oversimplified, as there are additional patterns that could be defined as states.

8. Scutellum - 0) without dark markings, or if any present they are restricted to extreme base, apex, or semicircular area on disc; 1) bicolored, with at least basal third on sides and dorsum distinctly darker than apex.

9. Mediotergite and/or subscutellum with lateral dark brown stripe or spot and body otherwise without dark markings, except sometimes on scuto-scutellar suture - 0) no; 1) yes.

10. Costal setulae of male wing - 0) slender; 1) stout, often spur-like.

11. Vein R₂₊₃ - 0) more or less straight; 1) sinuate; 2) with sharp bends, often with spur veins arising from them.

12. Vein M distal to DM-Cu - 0) straight or posteriorly curved; 1) anteriorly curved in distal half, but meeting costa at distinct angle; 2) more strongly curved apically, meeting costa in a smooth curve; 3) extremely strongly curved. Because there is nearly continuous variation in this character, it was difficult to divide into states and to code for many taxa; future refinement and rechecking of this character is needed.

13. Wing pattern - 0) not Anastrepha-type nor wasp mimic-type; 1) Anastrepha-type, with C-, S- and V-bands and hyaline area at apex of vein R₁; 2) Anastrepha-type, but without hyaline area at apex of vein R₁ (with complete costal band as well as S-band); 3) wasp mimic-type, with only broad complete costal band and faint streak over cell bcu.

14. S-band with basal cleft - 0) weak or absent; 1) strong.

15. Abdomen - 0) not petiolate; 1) petiolate.

16. Abdomen - 0) unicolorous except paler posterior margins of tergites; 1) bicolored, partially dark brown. The dark markings occur in various patterns, some of which probably are not homologous.

17. Lateral surstylus length - 0) elongate, apically with anterior and posterior lobes; 1) short to medium length (shorter than epandrium height), without anterior or posterior lobes but not transversely flattened; 2) short to medium length, oriented obliquely; 3) short to medium length, apically transversely flattened; 4) extremely short, barely extended beyond prensisetae.

18. Lateral surstylus - 0) not bootshaped; 1) short and somewhat bootshaped, with laterally projecting apical lobe.

19. Proctiger - 0) weakly or not creased, sclerotized area usually continuous; 1) with pair of strong lateral creases, sclerotized area divided into 3 parts.

20. Phallus length - 0) more than 1.25 mm long, glans present; 1) short, less than 1.25 mm long, glans absent; 2) less than 0.30 mm long, glans absent.

21. Glans - 0) strongly sclerotized basally and medially, subapical lobe not T-shaped if present; 1) weakly sclerotized medially, subapical lobe T-shaped.

22. Glans - 0) without spines; 1) with minute spines.

23. Oviscape basally with lateral, flange-like lobes - 0) no; 1) yes.

24. Eversible membrane enlarged basally and bearing group of enlarged dorsal teeth - 0) no; 1) yes.

25. Eversible membrane dorsobasal teeth pattern - 0) usually triangular or semicircular, with teeth well sclerotized and gradually changing in size; 1) all teeth relatively small and weakly sclerotized; 2) small and weakly sclerotized except for medially interrupted apical row of large, strongly sclerotized, hook-like teeth; 3) teeth short and arranged in elongate pattern.

26. Aculeus width - 0) greater than 0.05 mm; 1) less than 0.05 mm. Reduction in aculeus width has also occurred in some unplaced Anastrepha species.

27. Aculeus tip - 0) well defined, inner margin on ventral side distinct; 1) poorly defined, inner margin on ventral side indistinct.

28. Aculeus tip tapered, then parallel-sided, then tapered - 0) no; 1) yes.

29. Aculeus tip - 0) width < 0.18 mm, or if broad, not blunt; 1) extremely broad (width > 0.18 mm), blunt.

30. Aculeus tip - 0) elongate and/or narrow; 1) short, broad, with numerous fine serrations extended beyond base; 2) short, broad, and triangular, with very large serrations extended to base.

31. Spermathecae - 0) moderately sclerotized; 1) weakly sclerotized; 2) membranous.

32. Third instar larva, hairs of posterior spiracle - 0) medium length to long, longer than width of spiracular opening; 1) very short, less than width of spiracular opening.

33. Egg, micropyle end - 0) without lobe; 1) with short lobe; 2) with long lobe.

34. Chromosome number (diploid) - 0) 12 in both sexes; 1) 8 in both sexes; 2) female 12, male 11.