Schaffneropsallus, Schuh, 2006

Schuh, Randall T., 2006, Revision, Phylogenetic, Biogeographic, And Host Analyses Of The Endemic Western North American Phymatopsallus Group, With The Description Of 9 New Genera And 15 New Species (Insecta: Hemiptera: Miridae: Phylinae), Bulletin of the American Museum of Natural History 2006 (301), pp. 1-115 : 99-111

Schaffneropsallus , new genus

Type species: Schaffneropsallus oaxacensis , new species.

DIAGNOSIS: Recognized by the relatively small size, average total length 2.53, the pale background coloration and orange markings on the dorsum, the uniformly fumose membrane with pale veins (fig. 3), the presence of a conspicuous tubercle with associated heavy setae of moderately length on the left side of the pygophore (fig. 40E, F), and the broadly lanceolate right paramere with an elongate, finely attenuated apex (fig. 41). Similar in dorsal aspect to some species of Platyscytus in the pale coloration and contrasting markings on the dorsum, but distinguished from Platyscytus by the head in that group being elongate and tapered ventrally. Distinguished from all Phymatopsallus -group taxa by the marmorate membrane of those taxa and the differences in the right paramere, that structure being blunt apically with a more or less distinct projection on either side in the Phymatopsallus group, but strongly attenuated and spinelike in Schaffneropsallus .

DESCRIPTION: Male: Relatively small, elongate, nearly parallel-sided, total length 2.41–2.65, length apex clypeus–cuneal fracture 1.75–1.87, width pronotum 0.75–0.87. COL- ORATION (fig. 3): Body and forewings pale, with orange markings, head with a distinct stripe on the inner margin of each eye, pronotum with three longitudinal stripes, scutellum with a median longitudinal stripe, and hemelytra with numerous irregularly placed small orange spots; eyes silvery; appendages pale, femora with a few black spots; tibial spines dark with black bases; membrane fumose, veins pale. SURFACE AND VESTITURE (fig. 3): Dorsal body surface smooth, impunctate, weakly shining. Dorsal vestiture of recumbent, silvery setae. STRUCTURE: Head: Weakly transverse, posterior margin of eyes contiguous with anterior margin of pronotum, frons at most barely protruding beyond anterior margin of eyes ( Fig. 40A ); posterior margin of vertex indistinct; eyes large, showing sexual dimorphism, occupying entire height of head in lateral view (fig. 40A), vertex relatively narrow (fig. 3); antennae inserted below midpoint of eye, insertion contiguous with eye (fig. 40A); antennal segment 2 cylindrical, of approximately same diameter as segment 1, distinctly longer than width of head (ratio 72:60); labium reaching to apex of hind trochanters. Thorax: Mesothoracic spiracle and metathoracic scent-efferent system as in figure 40B, C. Legs: Claws smoothly curving over entire length, pulvilli small flaplike, arising from near midpoint of claw, parempodia setiform (fig. 40D). Abdomen: Tapered somewhat posteriorly. GENITALIA (figs. 40E, F, 41): Pygophore: Weakly conical, with a conspicuous tubercle on left side bearing heavy setae of moderate length. Vesica: Sshaped and three-dimensional, formed of two straps, apically with three slender spines, two extending in parallel well beyond gonopore, the other somewhat shorter and bent; secondary gonopore moderately large, well sclerotized, ovoid, without gonopore sclerite. Phallotheca: Elongate apically, relatively broad, with a sclerotized keel on apicodorsal margin, lacking spinelike projections as found in most Phymatopsallus -group taxa. Parameres: Left paramere with anterior and posterior processes simple, lacking in unique structural features; right paramere broadly lanceolate, apex elongate, slender, attenuated in the form of a sclerotized decurved spine.

Female: Elongate ovoid; total length 2.54– 2.63, length apex clypeus–cuneal fracture 1.77–1.86, width pronotum 0.75–0.84. COL- ORATION (fig. 3): As in male. SURFACE AND VESTITURE: As in male. STRUC- TURE: Hemelytra extending somewhat past apex of abdomen; eyes usually smaller than in male, vertex relatively broader (figs. 3); antennal segment 2 distinctly tapered proximally, in contrast to condition in male. GENITALIA (fig. 41): Sclerotized rings large, twisted; vestibulum forming a sclerotized, relatively short, tubular structure; posterior wall apparently simple with no visible ornamentation.

ETYMOLOGY: Named for Joseph C. Schaffner, in recognition of his work on the Mexican Miridae during his professorial career at Texas A&M University, in combination with the generic name Psallus . Gender masculine.

DISCUSSION: The spotting on the dorsum, albeit not uniform, and the presence of a conspicuous tubercle on the left side of the pygophore suggest a possible relationship of this taxon with the Phymatopsallus group. However, the membrane is not marmorate and the right paramere is lanceolate, drawn to a fine, elongate, sclerotized spine apically, and does not lay directly over the phallotheca in repose. The presence of a tubercle on the pygophore shows little congruence with other characters and is therefore not an unequivocal diagnostic feature for the Phymatopsallus group of genera.

In addition to Schaffneropsallus oaxacensis , new species, I have examined other material from southern Mexico in which the right paramere is almost identical in form, with the strongly attenuated spinelike apex. These specimens, representing more than one species, are all spotted, but do not possess the tubercle on the pygophore. Neither is the membrane marmorate, as in all known species here placed in the Phymatopsallus group.

Schaffneropsallus oaxacensis , new species figures 3, 40, 41; map 8

DIAGNOSIS: See generic diagnosis.

DESCRIPTION: See generic description.

HOST: Unknown.

DISTRIBUTION (map 8): Oaxaca, Mexico.

ETYMOLOGY: Named for its occurrence in the Mexican state of Oaxaca.

HOLOTYPE: MEXICO: Oaxaca: Jalapa del Marquez [16.5 ° N 95.46666 ° W], 04 Aug 1980, Schaffner, Weaver , Friedlander , 1 Oi ( AMNH _ PBI 00058285 ) ( TAMU).

PARATYPES: MEXICO: Oaxaca: 12.4 mi W of Tehuantepec, 16.33325 ° N 95.42124 ° W, 04 Aug 1980, Schaffner, Weaver, Friedlander, 1Oi ( AMNH _PBI 00058288) ( TAMU). Jalapa del Marquez, 16.5 ° N 95.46666 ° W, 04 Aug 1980, Schaffner, Weaver, Friedlander, 2Oi ( AMNH _ PBI 00058286, AMNH _PBI 00058287), 3♀ ( AMNH _PBI 00058289– AMNH _PBI 00058291) ( TAMU).

PHYLOGENETIC ANALYSIS

The Phymatopsallus group of genera offers a wealth of morphological information, particularly in the male and female genitalia, providing the potential for detailed phylogenetic analysis. Table 2 presents a list of character descriptions for the group; table 3 shows the distributions of those characters within the Phymatopsallus group, as well as three outgroup taxa.

The data shown in tables 2 and 3 were analyzed with NONA ( Goloboff, 1998) using the mult* (with 20 interations) and max* branch-swapping commands; the hold command was set at 10,000, allowing space for a maximum of 10,000 trees. The number of trees found was 127 with a length of 121 steps, consistency index of 55, and a retention index of 81. The strict consensus of those 127 trees resolves the genera recognized in the present paper but is not particularly informative with regard to the relationships among those genera. However, successive approximations weighting ( Farris, 1969; Carpenter, 1988) selected only one of the original trees, shown in figure 42; unambiguous optimizations of character data, as provided by the program WinClada ( Nixon, 2000), are shown. Because some characters allow for more than one possible optimization on the tree, these are not shown on the cladogram. Analysis of the same data with PIWE ( Goloboff, 1993, 1997)—which produces best fit trees by maximizing the sum of the average unit consistency index—using the same set of commands as above, produced a single tree shown in figure 42, with a fit statistic of 210.5.

Thus, although many of the characters in the dataset are homoplastic to a greater or lesser degree, as can be appreciated from the consistency index of 0.55, there is substantial phylogenetic signal from those characters as reflected in the retention index of 0.81 and the identical results produced through the use of successive approximations weighting of

TABLE 2 Descriptions of Characters and States TABLE 3 Character Data for Phymatopsalles-Group Taxa

the 127 most parsimonious trees found by NONA and the single tree found by PIWE.

Major phyletic lines within the Phymatopsallus group, indicated by node numbers in figure 42, are defined in part on the basis of the following characters:

Node 1. Phymatopsallus group: Membrane marmorate, right paramere moderately elongate, weakly parallel-sided, and with two terminal processes.

Node 2. Phymatopsallus clade: Apex of vesica with a single process about the length of secondary gonopore and phallotheca with a single thumblike process (‘‘spine’’) on the anterior surface.

Node 3. Stictopsallus clade: Vesical spine present and apex of vesica flattened.

Node 4. Salicopsallus clade: Pygophore with an elevated tubercle on the left side, tubercle with elongate simple setae.

Node 5. Salicopsallus : Dorsal margin of phallotheca with an elevated, sclerotized keel along entire length.

Node 6. Phymatopsallus: Vesica tubular, formed of a single strap, left paramere with thumblike process on anterodorsal margin, and dorsal margin of phallotheca beset with sclerotized tubercles, among other characters.

Node 7. Cercocarpopsallus clade: Antennal segment 2 not sexually dimorphic, and right paramere elongate, conspicuously parallelsided.

Node 8. Cercocarpopsallus: Pygophore with short, spinelike setae, triangular fold on dorsal margin of phallotheca.

Node 9. Angelopsallus clade: Apex of vesica a single process about 2 times length of secondary gonopore, phallotheca with a single spine on ventral margin, and vestibulum elongated with nautiloid enlargement distally and adjacent entrance-exit.

Node 10. Ceratopsallus clade: Pygophore with an elevated tubercle on left side, tubercle with elongate simple setae, left paramere with anterior process bifid, with posterior process curving, with terminal knoblike ornamentation, and distinctly separated from paramere body.

Node 11. Bisulcopsallus: Vesica very long, filamentous, left paramere with anterior process cloven, and vestibulum very long with a small nautiloid development.

Node 12. Ceratopsallus: Left paramere with anterior process cow horn-shaped, vestibulum with sclerotized envelope.

Whereas Knight (1964) conceived the tubercle on the pygophore (character 9) as a defining feature of Phymatopsallus as he construed it, the present phylogenetic analysis treats the tubercle as having evolved three times: in Schaffneropsallus , at node 4, and at node 10. Nonetheless, the tree shown in figure 42 maximizes agreement among a large number of characters on the tree and, as discussed further below, receives substantial ‘‘reciprocal illumination’’ ( Hennig, 1966) from patterns of geographical distribution and host associations.

BIOGEOGRAPHIC RELATIONSHIPS

Many phyline genera in the Nearctic, including, among others, Megalopsallus Knight (see Schuh, 2000a), Oligotylus Slater and Knight (see Schuh, 2000b), and Tuxedo Schuh (see Schuh, 2004a), as well as orthotyline genera such as Pseudopsallus Van Duzee ( Stonedahl and Schwartz, 1986) , Ramentomiris Stonedahl and Schuh, and Squamacoris Knight ( Stonedahl and Schuh, 1986) , are restricted to the western North America and show great diversity there. The patterns seen in the above-mentioned genera find parallels in other taxa such as Chlamydatus Curtis (see Schuh and Schwartz, 2005), Europiella Reuter (see Schuh, 2004b), and other groups of Phylinae , which, although not restricted to western North America, show their greatest diversity there and manifest restricted patterns of endemism. Phymatopsallus -group taxa show strong similarity to Megalopsallus in having substantial diversity in the Sonoran Desert, but unlike the former, are not primarily associated with halophytes.

In the broadest sense, Phymatopsallus - group taxa display distributions that might be classified as Sonoran–Chihuahuan, as seen in the Phymatopsallus clade, Californian, as seen in the genus Ceratopsallus , and Great Basin–Rocky Mountain, as seen in some members of the genus Bisulcopsallus . This classification, however, reflects the distributions of larger clades and some widespread species, and as such it is too crude to allow for the recognition of areas of endemism. As a way of conducting a more in-depth biogeographic analysis, the following areas of endemism were postulated on the basis of known Phymatopsallus -group distributions:

Southern Mexico (Oaxaca – Chiapas)

Southern Arizona + Sonora, Mexico

Imperial Valley (low elevation SE California)

Baja California (non-Mediterranean)

Southern New Mexico + west Texas

Southern Nevada

Southern California (San Joaquin Valley and adjacent Sierra Nevada)

Northern California (Sacramento Valley, adjacent Sierra Nevada, extreme SW Oregon)

Oregon (central and northern interior)

Utah + NE Arizona (montane areas)

Colorado + NE Arizona (montane areas)

Some of these areas, such as Southern New Mexico + West Texas and Oregon, are not unique to any single taxon, but nonetheless are not part of the intersection of distributions for two or more widespread species. Other recognized areas, such as northern California and southern California, might be viewed as overly crude characterizations, but existing distributional data, particularly in California, do not exist in sufficient detail to allow for more precise circumscriptions.

Phymatopsallus -group distributions were analyzed by assigning individual species to one or more of the areas identified above and then applying the methods described by Nelson and Ladiges (1996) for the construction of ‘‘paralogy-free’’ area cladograms. The result included seven subtrees, shown in the lower part of figure 43. Those subtrees are compatible and form the combined area cladogram shown at the top of figure 43.

Examination of the subtrees and combined area cladogram may allow us to draw some initial conclusions:

1. Southern Mexico is the uncontradicted sister area of the remaining areas on the cladogram.

2. Baja California is then sister to all remaining areas in those subtrees where it occurs and where there is resolution of areas beyond the basal dichotomy.

3. In all but one subtree (4), southern Arizona is then sister to all remaining areas in those subtrees where it occurs and where there is resolution of areas beyond the basal dichotomy.

4. Southern California and northern California appear to be sister areas.

5. Imperial Valley and Oregon are placed in relatively basal positions on the area cladogram because these areas are not distinct to any taxon, but rather part of more widespead distributions and—particularly in the case of Oregon —taxon distributions that include several other areas.

6. Colorado + E Arizona pertains only to Bisulcopsallus polhemorum , whose position on the cladogram is unresolved relative to other members of the genus. For this reason, and because this area is unique to B. polhemorum , its position in the area cladogram is also ambiguous.

With these observations in hand, we might conclude that although some area relationships appear to be revealed by Phymatopsallus -group taxa, others remain ambiguous because of widespread distributions and lack of resolution in certain areas of the taxon cladogram. What seems certain is that the Sonoran Desert region of Arizona is extremely diverse within the taxon, and that it forms a complex relationship with both California and areas in the interior of western North America. The latter aspect of this relationship is seen graphically in maps 2, 3, and 4, where Bisulcopsallus is restricted to the Rocky Mountain system, western Texas, and higher elevations in Arizona, whereas its sister group Ceratopsallus is diverse and widespread in California, with limited occurrences in adjacent southern Nevada and Arizona, and overlapping with Bisulcopsallus only in those areas.

It is equally clear that knowledge of diversity in the group in Mexico is woefully inadequate compared to that for the United States. In the long run, the ranges of some additional species will certainly be shown to extend into northern Mexico, with the possibility that additional diversity will be discovered there.

Even in the face of inadequate geographic sampling, the distributional patterns in the Phymatopsallus group find certain broad parallels in what we see in other taxa. Notable among these is the sister-area relationship of portions of central/southern Mexico with the western United States and adjacent Mexico. This pattern has been documented in phylogenetic analyses of the Squamocoris generic group by Stonedahl and Schuh (1986) and the Pseudopsallus generic group by Stonedahl and Schwartz (1986, 1988).

Furthermore, the virtual restriction of certain clades such as Ceratopsallus to California and northern Baja California, with some outlying taxa or distributions in Arizona, is repeated in many other genera, but Oligotylus Van Duzee , Phallospinophylus Weirauch , Pygovepres Weirauch , and Tuxedo Schuh might be cited as excellent examples.

Additional fieldwork and analytic work on the faunas of Mexico and the American West should help to clarify the generality of these patterns as well as better define areas of endemism.

HOST RELATIONSHIPS

To better understand the evolution of host associations in Phymatopsallus -group taxa, host occurrences have been optimized on the cladogram of bug relationships shown in figure 42, following the approach taken by Carpenter (1989). Host occurrences were treated as states of a character and optimized on the morphology-based cladogram. Host data were coded at the family level, and thus no distinction was made between occurrences on Acacia , Cercidium , and Prosopis (Fabaceae) , Ceanothus , Rhamnus , and Zizyphus (Rhamnaceae) , or Amelanchier , Cercocarpus , and Purshia (Rosaceae) . Ceratopsallus pintoi and C. quercicola were coded as polymorphic because they have recorded hosts from different plant families. Single occurrences on a given host were sometimes not included when contradicted by all other data. Figure 44 reproduces the topology of figure 42 but replaces bug taxon names with host occurrences. Optimization of hosts by num- bered nodes and terminal taxa on the cladogram is according to the pattern on p. 110.

The basal host condition for the Phymatopsallus group is ambiguous because the host of the immediate outgroup, Schaffneropsallus , is unknown and because available character data do not form a dichotomous resolution at the base of the tree. Nonetheless, three family-level host affiliations are evident within the analysis.

First, there is strong evidence for the original restriction of Phymatopsallus clade (node 2) to the Fabaceae , including particularly the genera Acacia , Cercidium , Psorothamnus , Olneya , and Prosopis . These woody genera are all dominant landscape elements in the American Southwest. The Salicaceae-Rhamnaceae feeding preferences within Salicopsallus (node 5) would appear to be secondarily derived. No other Phymatopsallus -group taxon is known to feed on the Fabaceae .

The Cercocarpopsallus clade (node 7) is basally associated with Cercocarpus (Rosaceae) , an association based on multiple records for Cercocarpopsallus bispinosus and a single record for C. gracilis , but with multiple collection events for the latter species, all of which are consistent with a Cercocarpus association. The subsequent branching event (node 9) is optimized to treat Quercus as the basal condition. Sideroxylon (Sapotaceae) , the host of Angelopsallus , is based on a single collection, and when optimized on the current cladogram appears as a unique event.

Optimization of node 10 continues to treat Quercus (Fagaceae) as basal within the taxa arising from that node. Although hosts are not known for two of the species of Bisulcopsallus (node 11), the basal condition in the taxon is optimized as Quercus with Fraxinus -feeding being treated as autapomorphic. This optimization allows the prediction that B. fulvipunctatus and B. pallidus are also Quercus feeders.

The basal condition for Ceratopsallus (node 12) is optimized as Ericaceae , with Fagaceae being a secondary reacquisition in the clade (node 13) containing C. croceus , C. pantherinus , and C. pintoi . Further collecting and host documentation for Ceratopsallus , as well as a more robust result concerning relationships within the group, will help corroborate or reject these conclusions.

Although Ceanothus , Cercocarpus , Quercus , and Salix , among other plant taxa, serve as hosts to Phymatopsallus -group species—as well as large numbers of other Miridae species in North America—the association of the bug taxa across plant lineages appears to be primarily ecological, rather than systematic. This theory receives support from prior studies in the Miridae , such as those of Stonedahl and Schwartz (1986) on Pseudopsallus and Stonedahl and Schuh (1986) in Squamocoris and its near relatives. In some cases, such as Plagiognathus Fieber , the evidence for ecologically derived host associations is particularly graphic, with multiple independent instances of Coniferae feeding embedded with a larger Angiospermae-feeding group of bugs ( Schuh, 2001).

Even though Phymatopsallus -group taxa feed on a relatively broad array of host lineages, the optimization of host occurrences on the Phymatopsallus -group cladogram (fig. 44) indicates substantial congruence between host preference and phylogenetic relationships within the bugs. This ‘‘reciprocal illumination’’ ( Hennig, 1966) demonstrates the power of phylogenetics in evolutionary biology, and helps to predict where we might best concentrate our efforts in the search for new host records for the Phymatopsallus group.

Figures

Fig. 40. Schaffneropsallus oaxacensis. A. Lateral view of head. B. Mesothoracic spiracle and metathoracic scent-efferent system. C. Detail of metathoracic scent-efferent evaporatory area. D. Frontolateral view of pretarsus. E. Lateral view of pygophore, tubercle at upper left. F. Detail of setae on tubercle of pygophore.

Fig. 44. Taxon-host cladogram for Phymatopsallus-group taxa (see text for explanation).