Half-jumping plant lice – a taxonomic revision of the distinctive psyllid genus Togepsylla Kuwayama with a reassessment of morphology (Hemiptera, Psylloidea)

Abstract Togepsyllinae is a distinctive group within Psylloidea, with its systematic status treated variously by different authors. Of the only two known genera, Togepsylla is better known and distributed in temperate and tropical East Asia. In this study, the taxonomy and morphology of Togepsylla was studied in detail. Keys to adults and fifth instar immatures of the genus are provided. Togepsylla glutinosae sp. n. is described, and T. matsumurana, T. takahashii, and T. tibetana are redescribed. Syncoptozus is compared with Togepsylla for differences in morphology. Modern psyllids have evolved their jumping hind legs via the elongation of the dorsal edge of coxa, the broadening of coxal wall, the thickening of meron, the backward twisting of the plane of trochanter, femur, and tibia, and the enlargement of trochanteral tendon. However, in Togepsylla, this modification has progressed halfway. The metapleuron of Togepsylla is arranged in a different way than other psyllids. The pleural sulcus is short, and the metepisternum and trochantin are not divided. Wax-secreting fields on abdominal sternites, resembling those of whiteflies, are found on all Togepsylla species, and described for the first time. Other distinctive characters of the genus are also revealed, e.g. frons completely fused with gena, a pair of extra sclerites present behind the base of thoracic furca, one-segmented aedeagus, and absence of a flag lobe on valvula dorsalis of ovipositor. Based on various similarities in morphology, Togepsyllinae may have a close relationship with Aphalaridae-Rhinocolinae and is possibly related to Homotomidae, Liviidae-Liviinae and Atmetocranium (Calophyidae). All the distinctive characters of Togepsyllinae suggest that the current placement of the group is doubtful, and the phylogeny of Aphalaridae needs to be resolved.


Introduction
Psylloidea, the superfamily of jumping plant lice, is a group of phloem-sucking insects with strict host specificity. Among all its subordinate taxa, the Oriental and Palaearctic genus Togepsylla is undoubtedly one of the most distinctive. Named after the symmetrical and articulated thick 'spines' on the dorsum of head and thorax, it is readily diagnosed by these spine-like setae and wings held flat over its dorsum instead of roofwise, in addition to the more intrinsic characters identified in this article. The genus contains only three known species to date: Togepsylla matsumurana Kuwayama, 1949, T. takahashii Kuwayama, 1931 (type species), and T. tibetana (Yang & Li, 1981). The few known species develop exclusively on Lauraceae plants, often inducing pit galls or leaf-rolling galls (Takahashi 1936;Miyatake 1970;Li 2011), in contrast to their spine-lacking Neotropical relative Syncoptozus Enderlein, which develops on plants of Magnoliaceae. The two genera compose the small taxon Togepsyllinae/-ini with valid monophyly. Yang and Li (1981) erected Hemipteripsylla to contain H. tibetana and T. matsumurana, which possess relatively short setae based on short projections and lack tiny spines on the dorsum of aedeagus. These characters were accurately recognized, although considered insufficient to support an alternative genus, by Hodkinson (1990), who synonymized Hemipteripsylla with Togepsylla.
Togepsyllini was established by Bekker-Migdisova (1973), who assigned it to her sense of Carsidaridae-Tenaphalarinae. Before Bekker-Migdisova, Kuwayama (1931) placed Togepsylla in Carsidarinae, whereas Miyatake (1970) moved the genus into Pauropsyllinae. White and Hodkinson (1985) presented the first calculated cladogram of Psylloidea, using a numerical method. They treated the group as Togepsyllinae, assigning it in Aphalaridae, one of the eight families recognized in their classification. The authors postulated an extinct Rutales-feeding ancestor of all modern psyllids; however, they also mentioned in the last sentence of that paragraph that "Togepsylla (Aphalaridae), which feeds on Lauraceae (Annonales), may possibly be a relic genus of a psyllid group which antedates the Rutales-feeders." On the other hand, using the now synonymized Hemipteripsylla as the type genus, Li (2011) elevated Togepsyllinae to superfamily level, Hemipteripsylloidea, containing a single family with three genera. He recognized Hemipteripsylloidea as a primitive member of 'Psyllidomorpha' without a supporting phylogenetic analysis, simply based on the flat-held wings and onesegmented instead of two-segmented aedeagus in males. In the most recent, influential classification of Burckhardt and Ouvrard (2012), Togepsyllinae was treated as one of the five subfamilies of Aphalaridae whose fifth-instar immatures lack an unguitractor on the tarsal arolium, closer related with Rhinocolinae and Spondyliaspidinae.
Based on a synthesis of fossil studies, the defining character of Psylloidea sensu stricto is the mutual modification of the metacoxa and furca of the metathorax (Ouvrard et al. 2010) to allow the attachment of jumping muscles (Bekker-Migdisova 1971), thus enabling psyllids to jump. According to limited observations, psyllids appears to do a mid-air somersault in jumping. The jumping of Psylla alni (Linnaeus), Cacopsylla peregrina (Foerster) and Psyllopsis fraxini (Linnaeus) is triggered by pressing hind trochanters and femora against the coxal wall, then a quick release of the tension, powered by the metathoracic internal muscles attached to the furca and trochanteral tendon (Burrows 2012). The release is strong enough to rotate the entire body upside down, but the fore legs provide a support to maintain balance, and then the entire insect is cast airborne and starts to rotate forward rapidly, sometimes with the wings opening in mid-air to commence flight. The "non-jumping plant lice" of the extinct families Liadopsyllidae and Malmopsyllidae, which date back to the Mesozoic and possess unmodified metacoxa, provides further support that the modified hind legs are monopolized by modern Psylloidea (Ouvrard et al. 2010). Occasionally, some extant psyllids, such as Togepsyllinae and Apsylla cistellata (Buckton 1896) (Aphalaridae: Rhinocolinae), are not known to jump. These exceptions are regarded by Ouvrard et al. (2010) as secondary reductions.
A new species, Togepsylla glutinosae sp. n., has been collected on Litsea glutinosa from tropical China recently. During the taxonomic work on Chinese Aphalaridae, many previously overlooked characters of Togepsylla were revealed, including the halfmodified metacoxae which do not provide sufficient jumping force, and wax secreting fields on abdominal sternites first found on a psyllid. Based on an integration of East Asian collections, this study aims to revise the taxonomy of the genus Togepsylla, and reassess the distinctive morphological characters of it.

Materials and methods
This study is based on the collections of the Entomological Museum of China Agricultural University (CAU), the Natural History Museum, London (BNHM), Hiromitsu Inoue's personal collection (HIC), and the Osaka Museum of Natural History (OMNH). The type series of Togepsylla takahashii and T. matsumurana, previously preserved in Hokkaido University, were lost according to Jin Hyung Kwon (personal communication). Specimens examined regarding the non-Togepsylla species involved in discussion are listed as follows:

BNHM
Atmetocranium myersi Ferris & Klyver, 1932 Slides were prepared following this protocol: whole insect soaked in boiling potassium hydroxide (KOH) solution for 10 minutes, naturally cooled down after heating stopped, washed in distilled water, and finally mounted on a slide in glycerine. All drawings and examinations were undertaken with an Olympus BX41 microscope. One dry-mounted adult of each Togepsylla species was coated with gold using a Leica EM SCD050 Super Cool Sputter Coater and prepared for Scanning Electron Microscope (SEM) examination and photography, using an FEI ® Quanta 450 Environmental Scanning Electron Microscope.
Measurements were taken with a Keyence VHX-1000 digital microscope using the measuring function and are given in millimeters (mm). For adults: HW head width, AL antennal length, SL length of the posterior pair of prickly setae on the median of the vertex, TW mesoscutum width, WL fore wing length, TL metatibial length.
For fifth instar immatures: BL total body length, AL antennal length, HW head width, FL fore wing pad length, measured as the distance between the transverse tangents of the anterior angle and posterior margin.
Total body length of adults was not determined because most dry-mounted specimens examined were in distinct positions and therefore incomparable.
Comparisons of morphological characters were based on direct observations of specimens, drawings, SEM photos, measurements, and sometimes on the literature. Putatively homologous characters were compared across the concerned taxa. Diagnostic characters are described for each species, with emphasis on the differences. Characters potentially useful for systematic studies are noted and are referred to in the discussion.
Scientific names of plants follow the Missouri Botanical Garden (2016), and higher systematics of plants (except for the direct citation of older literatures) follow The Angiosperm Phylogeny Group (2016).

Togepsylla Kuwayama, 1931
Togepsylla Kuwayama, 1931: 121 Head slightly inclined from longitudinal body axis. Vertex lacking median suture; two tubercles present along the median line, each bearing a pair of prickly setae. Base of lateral ocelli moderately bulging, each bearing two prickly setae. Vertex consistent with gena. Plane of torulus about perpendicular to that of the vertex. Frons completely fused with vertex and gena, only moderately raised from the surface. Gena not divided into two lateral parts, but firmly compact as one, with roughly symmetrical simple setae (Fig. 47); parts below torulus sometimes produced. Occiput smoothly connected with vertex, not folded below it. Plane of postocular sclerite about perpendicular to that of vertex, not nearly parallel with it. Antennae 10-segmented, surface sculptured with minute spinules arranged in transverse rows; at least six rhinaria present on apices of segments IV-IX, apex of segment III sometimes also with one; segments IV, VI and VIII sometimes possess extra rhinaria; rhinarium with closely packed minute spinules lining below, usually bearing horn-like projections. Clypeus rather short, with no extra seta except the apical pair of setae. Labium rather short, two-segmented, lacking 'conical sensoria' (as termed by Liang et al. 2013) on the tip (Fig. 48).
Legs long and slender. The three sensory pores on femora ventrum arranged in a row. Plane of hind legs almost parallel with that of middle legs. Metacoxa with rather large tubercle above apical opening, and lacking meracanthus (Fig. 56). Metafemur without a cluster of thick setae on the outside of apex. Metatibia without genual spine, with 1-3 rows of thick setae; apical part often with a row of tightly packed setae dorsally; apical spurs relatively long and slender, sclerotized at different extents but never reaching the hard and black status as in most other psyllids, forming an open crown. Metabasitarsus lacking sclerotized spurs on the apex, but with one or two (in other psyllids there is only one) pairs of simple setae. Apical tarsus with a pair of short and tapered apical setae. Claws with rounded or rather narrow pulvilli.
Fore wing narrowest in the base and gradually becoming much wider apically, usually widest at subapex or apical 1/4. Costal break present. Pterostigma absent. Vein Rs reaching anterior margin instead of apical margin. Cell cu 1 rather long and flat. Veins A 1 and A 2 touching in the middle. Anal break adjacent to the apex of vein Cu 1b.
Hind wing with partially thickened anterior margin. Veins A 1 and A 2 combined or one is lost (probably A 1 ), leaving a thickened vein A; cell a 1 lost (Fig. 49).
Tergite of abdominal segment 1 better developed, with a median sclerite present (Fig. 50). Spiracles of segments 1 and 2 invisible. Sternites of segments 4-6 each with a pair of wax-secreting pore fields laterally, with shape variable.
Male terminalia: In natural status, proctiger, aedeagus and parameres all oriented caudally instead of upwards. Posterior aspect of proctiger enveloped. Aedeagus unisegmented and simple, sometimes with tiny spines on dorsum. Sperm pump with only basal end plate, lacking apical end plate (Fig. 51).
Female terminalia: Subgenital plate placed much more proximal than proctiger and simple, lacking tip sometimes. Proctiger lacking rows of long setae on the dorsum. Valvula dorsalis of ovipositor without flag lobe. Median valve slender and placed more terminal, apex touching the subapex of ovipositor.

Key to adults of Togepsylla
Mesoscutum with four pairs of prickly setae. Metatibia with three short rows of thick setae, lacking a tightly packed row of short setae on the dorsum (Fig. 13). Apex of metabasitarsus with only one pair of simple setae (Fig. 13). Pulvilli broadly rounded (Fig. 13). Fore wing cell cu 1 tallest in apical 1/3, with vein Cu 1a abruptly curved at the point; vein M 3+4 completely decorated with setae; surface spinules absent; fields of radular spinules relatively large (Fig. 9).
Pore fields on abdominal ventrum small oval; pores loosely packed (Fig. 16). Male terminalia: Proctiger slightly curved backwards (Fig. 19). Paramere small lamellar; apical half of anterior margin with a thin lobe stretching inwards; anterior margin of basal 1/3 emarginated; two long and thick setae present on inner surface, near the anterior margin; inner surface with a curved ridge decorated with thick setae on apical half (Figs 19-21). Aedeagus with a few tiny spines on the dorsum (Fig. 19). Base of subgenital plate with a small cluster of setae (Fig. 19).
Female terminalia (Fig. 29): Oblong in overall shape. Base of proctiger slight raised, anus partly sunken; dorsal view of proctiger constricted at apical 1/3; apical process with small amounts of tiny setae. Subgenital plate lacking tip, with sparse setae on ventral surface.

Togepsylla matsumurana
Redescription. Adult coloration. Head yellow, vertex with brown patterns. Long and thick setae on dorsum black. Compound eyes light brown. Ocelli yellow. Antennae yellow, segments I-II light brown, apices of segments III, IV, VI, VIII black, segments IX-X entirely black. Thoracic dorsum brown, except for bases of setae which are yellow. Thoracic pleurites light brown. Legs yellow, with apical half of femora light brown, apex of tibiae brown. Fore wing membrane hyaline and colorless; R 1 , apices of Rs and M 1+2 black (Fig. 10). Abdominal tergites of segments 1-5 black, sternites brown. Male proctiger brown. Female terminalia yellow.
Structures: Setae on dorsum of body relatively short (Table 1) and based on smooth projections. Vertex with a pair of small foveae between median-posterior tubercle and lateral ocelli (Fig. 2). A pair of small tubercles present above toruli (Fig. 2). Genal tubercles strongly protruding (Fig. 2). Antennal segments IV-IX each with a single rhinarium on apex, segment IV with one, segments VI and VIII each with two extra rhinaria; rhinaria without horn-shaped projection; proximally based terminal seta slightly longer than the distally based one (Fig. 6).
Mesoscutum with 5 pairs of prickly setae (Fig. 52). Metatibia with one row of thick setae ventrally, and with a tightly packed row of long setae on the dorsum (Fig.  14). Pulvilli narrow (Fig. 14). Fore wing with broad cell r 1 , cell cu 1 tallest in the middle; vein M 3+4 without seta; surface spinules rather minute, widely spread across a relatively small area in distal cells; fields of radular spinules relatively large (Fig. 10).
Pore fields on abdominal ventrum large oval, with pores loosely packed (Fig. 17). Male terminalia: Proctiger slightly curved backwards apically (Fig. 22). Paramere broad lamellar, with rather slender base; anterior margin of apical half emarginated and thin; posterior margin with a basal ridge on outer surface; middle of inner surface with a small area of netlike grains; anterior angle with a few short and thick setae on inner surface; posterior margin with a band of inner-curved short setae on apical 2/3 (Figs  22, 23). Tip of aedeagus forming an acute small hook, dorsum of aedeagus lacking tiny spines (Fig. 24). Subgenital plate with moderately produced dorsal-apical angle, and with a few setae on the base (Fig. 22).
Female terminalia (Fig. 30): Short and broad in overall shape. Apical 1/3 of proctiger strongly curved upwards; apical half of proctiger with nearly evenly spaced setae, and with a row of setae along ventral margin of apical process. Subgenital plate with blunt and retracted apex, ventral surface with sparse setae.
Redescription. Adult coloration. Ground color yellow. Compound eyes grey. Long and thick setae on dorsum black. Ocelli yellow. Antennae yellow, with black spices on segments III-VIII; segments IX-X entirely black. Fore wing hyaline, with four obliquely transverse yellow stripes (Fig. 11). Legs yellow. Abdominal tergites brown. Male and female terminalia yellow.
Structures: Setae on dorsum of body relatively long (Table 1) and based on prominent projections. Torulus produced and slightly turned outwards (Fig. 3). Gena flat (Fig. 3). Antennal segments III-IX each with a single rhinarium on the apex, the ones on segments V and VII with small horn-shaped projections; proximally based terminal seta slightly longer than the distally based one (Fig. 7).
Mesoscutum with four pairs of prickly setae (Fig. 53). Metatibia with two rows of thick setae lateral-ventrally, and with a tightly packed row of short setae on the dorsum (Fig. 15). Apex of metabasitarsus with a pair of thick setae (Fig. 15). Pulvilli narrow (Fig. 15). Fore wing with long and narrow cell r 1 , vein M 1+2 rather close to vein Rs, cell cu 1 tallest in the middle; vein M 3+4 lacking seta; surface spinules as tiny thick spines, widely spread across a large area on wing membrane; fields of radular spinules relatively large (Fig. 11).
Male terminalia: Distal 1/3 of proctiger with posterior surface split and replaced with membranous tissue (Fig. 25). Paramere slender and bilobed; apex of anterior lobe developed into a sclerotized tooth; anterior margin of basal 1/3 emarginated; two long and thick setae present on inner surface, near the anterior margin; apical half with a curved vertical row of small peg setae on inner surface, near posterior margin (Fig. 25,  26). Aedeagus curved forward at apical 1/4, dorsum with a short row of spines that gradually turn smaller apically (Fig. 25). Dorsal-apical angle of subgenital plate produced and with a long seta (Fig. 25).  Female terminalia (Fig. 31): Long and straight in overall shape. Base of proctiger slight raised, apical process without tiny setae. Subgenital plate with acute apex, ventral surface with relatively dense and nearly evenly spaced setae.
Redescription. Adult coloration. Ground color yellow. Long and thick setae on dorsum yellow. Compound eyes grey. Ocelli yellow. Antennae yellow, with black spices on segments IV, VI, VIII; segments IX-X entirely black. Fore wing hyaline and colorless (Fig. 12). Male and female terminalia yellow.
Structures: Setae on dorsum of body relatively short (Table 1) and based on smooth projections. A pair of small tubercles present above toruli (Fig. 4). Gena moderately swollen bilaterally (Fig. 4). Antennal segments IV-IX each with a single rhinarium on the apex, segments IV, VI and VIII each with an extra rhinarium; all rhinaria with horn-shaped projections; proximally based terminal seta about equally long with the distally based one (Fig. 8).
Mesoscutum with five pairs of prickly setae. Metatibia with one row of thick setae ventrally, and with a tightly packed row of long setae on the dorsum. Pulvilli narrow. Fore wing with broad cell r 1 , cell cu 1 tallest in the middle; vein M 3+4 with one seta on the base; surface spinules rather minute, widely spread across a large area in distal cells; fields of radular spinules unclear (Fig. 12).
Pore fields on abdominal ventrum large oval, with pores loosely packed. Male terminalia: Proctiger completely sealed, with apex slightly thickened (Fig. 27). Paramere broad lamellar, with rather slender base; anterior margin of apical half emarginated, thin and serated; posterior margin with a basal ridge; apical half of inner surface with netlike grains; anterior angle with a few short and thick setae on inner surface; posterior margin with a row of inner-curved short setae on apical half (Figs 27, 28). Aedeagus curved backwards apically, dorsum lacking tiny spines, tip forming a small acute hook (Fig. 27). Subgenital plate near rectangular in profile, dorsal-apical angle with one long seta, ventral surface with sparse setae (Fig. 27).
Female terminalia (Fig. 32): Short and broad in overall shape. Apex of proctiger moderately curved upwards; apical half of proctiger with nearly evenly spaced setae, and with a row of setae along ventral margin of apical process. Subgenital plate with blunt and retracted apex, ventral surface with sparse setae. Biology. Yang and Li (1981) recorded that the adults gather among the clusters of young leaves by large amount. The record of a habit similar with T. takahashii by Li (2011) seems artificial.

Differences between Togepsylla and Syncoptozus
The similarities and differences of the two genera have been listed by Hodkinson (1990). Nevertheless, some supplements can still be made here. Togepsylla possesses no median suture or discal foveae on the vertex; while Syncoptozus has the anterior section of median suture present, and S. bifurcatus possesses discal foveae (Brown and Hodkinson 1988). Togepsylla has rhinaria on antennal segments IV-IX, sometimes even segment III, and often with additional rhinaria; Syncoptozus has only one rhinarium on apex of segments IV, VI, VIII, and IX each.

Hind legs
Psyllids jump powerfully, then cast a mid-air rotation. Such a somersault, however, involves not only the strong muscles supported by the specialized metathoracic furca, enlarged metatrochanteral tendon and expanded meral part of the metacoxa but also a kicking of both hind legs on parallel planes (Burrows 2012), which are also parallel to the longitudinal body axis. This longitudinal placement of hind legs is caused by an inward twist of the metacoxa.
To discuss the formation of the enlarged and twisted metacoxa, one must seek reference from the mesocoxa. Mid and hind legs are both appendages of winged thoracic segments; additionally, in immature psyllids, they are equal in every detail, although differing from the forelegs in some aspects, indicating that hind legs of adults emerged from the model of mid legs. An undescribed Cacopsylla species is used as example: The mesocoxa (Fig. 54) are relatively small and are connected to the coxal condyle of the mesopleurite by a dorsal-most articulation. Starting from the articulation, a thickened vertical edge runs down the outer surface, facing the lateral aspect, and is termed here as the 'dorsal edge' of the coxa. The coxa connects to the trochanter via two 'trochanteral condyles', which are longitudinally positioned, thus respectively termed 'anterior-' and 'posterior trochanteral condyle'. Such longitudinal positioning of trochanteral condyles places the mid legs on a transverse plane, a plane nearly perpendicular to the longitudinal body axis. Besides, a normally developed trochanteral tendon originates on the inner-dorsal edge of the mesotrochanter, stretching into the chamber of the mesocoxa, clinging onto the corresponding muscles.
Compared with mesocoxa, the metacoxa (Fig. 55) first experienced an enlargement of the coxal wall, which pivots over the elongation of the dorsal edge and is primarily characterized as the expansion of the prearticular part of the coxal wall and thickening of the meron. Simultaneously, because of the unequal development of the prearticular part and the meron, the entire metacoxa is twisted backwards at approximately 90°, turning the two trochanteral condyles into a transverse position. The plane of hind leg is therefore turned longitudinal (Figs 66, 67). This pair of straightly backwards-reaching hind legs provides a much better concentration of jumping force, thereby driving the powerful jump described above. Additionally, the trochanteral tendon is magnified and possesses a tortuous apex, serving to support the strong jumping muscles.
By contrast, Togepsylla possesses half-modified metacoxae (Fig. 56). The enlargement is almost complete, but the positioning of the two trochanteral condyles is shifted at a limited level. For this reason, the hind legs of Togepsylla retain a posture similar to that of the middle legs, as shown in the habitus photograph (Fig. 64). Additionally, the trochanteral tendon is also half-enlarged: the relative size is much smaller, and the apex, although also expanded, is a simple flat surface instead of tortuous. According to the field observations by Xinyu Luo, adults of Togepsylla glutinosae sp. n. can only leap forward like frogs, at a short distance and without mid-air rotations.

Lateral aspect of thorax
Most psyllids possess an apophysis on meso-and metepisternal complex, termed 'trochantinal apodeme' (Ouvrard et al. 2002). This is an autapomorphy of Psylloidea. For mesopleuron, this structure may be on the anterior margin or median portion, depending on the taxon (Ouvrard et al. 2002). However, there are some cases like Togepsylla and Pseudophacopteron in which the trochantinal apodeme is placed on the anterior margin and reduced to an obscure vestige.
According to Ouvrard et al. (2002), the modification of psyllid metapleurite relative to mesopleurite is due to a curving of the pleural sulcus. For the metapleurite of most psyllids, taking Cacopsylla as example, the pleural sulcus turns downwards over the coxal condyle, becoming congruent with the elongated and internally ridged dividing suture of episternum and trochantin (Fig. 58). In Togepsylla, the metapleuron represents a halfway modification. The dividing suture of metepisternum and trochantin is absent, the trochantinal apodeme is shallow and in anterior position, as in mesothorax (Fig. 57).

Wax-secreting fields on abdominal sternites
Togepsylla possesses three pairs of fields of pores on sternites of abdominal segments 4-6, in both sexes. Wax secretions from these pores have been observed on T. matsumurana (Fig. 62). Similar structures, several pairs of wax plates composed of many small wax-secreting pores, is one of the defining characters of adult whiteflies: Whiteflies kick the wax secretions of these glands with the hind legs, and then spread the shattered wax particles over the entire body surface (Byrne and Bellows 1991). In females of Aleyrodinae, two pairs of wax plates are found, on segments 3-4, whereas in Aleurodicinae, four, on segments 3-6; in males of Aleyrodinae, four pairs are present on segments 3-6, whereas three pairs appear on segments 3-5 in Aleurodicinae and Udamoselinae (Gill 1990;Martin 2007).
All the four members of Sternorrhyncha are known to secrete wax through integumental wax gland/pores. In scale insects whose wax glands are studied the most, these structures are highly variable in ultrastructure (shape and number of loculars of each pore) and distribution (all over the body or restricted to a certain region) (Foldi and Pearce 1985;Foldi and Lambdin 1995). Some aphid families/subfamilies possess wax gland plates, which also vary in shape and distribution, on body dorsum (Chen and Qiao 2012). These, however, are not so far known to reveal the same arrangement as Togepsyllinae and whiteflies, nor does the lack of detailed ultrastructural study of Togepsyllinae support their resemblance.
Psyllid immatures possess wax-secreting pores on their caudal plates. These pores are arranged in various patterns, mostly with a basic circum-anal ring (possibly homologous with the circum-anal ring of female adults), and on many occasions with extra pore fields (Brown and Hodlinson 1985). Extra pore fields can sometimes be succeeded by the adults, appearing on their more terminal (usually segments 7 and/or 8) abdominal tergites, e.g. Agonoscena pegani Loginova, 1960and A. sabulisa Li, 1994(in Li et al. 1994) (Luo 2016). Although it is not currently possible to accurately decide the homology between abdominal segments between immatures and adults, one can still roughly judge and count the separate segments of immatures by the dorsal and ventral setae rows. So far, the immature of not any species possess wax secreting pores on areas that are possibly homologous with abdominal sternites 4-6. This is the first time that a psyllid adult is found with such fields of wax-secreting pores. Compared with those of whiteflies, wax pore fields of Togepsylla are strongly constricted, and the segment correspondence is different. It is uncertain whether these structures of Togepsylla and Aleyrodoidea are homologous or not.

Discussion
Togepsyllinae displays great differences with all the other psyllid taxa in external morphology. These include: frons completely fused with gena; gena firmly compact instead of being bisected; 'conical sensoria' absent from apex of labium; metapleuron distinctively arranged; metacoxa ventral aspect of metathorax as a compact sclerite; wax plates present on sternites of abdominal segments 4-6; male terminalia oriented caudally; male proctiger completely enveloped, instead of having a basal major part, which is sclerotized anteriorly and laterally, whereas membranous posteriorly, with a median suture; aedeagus one-segmented; sperm pump with only basal end plate, lacking the apical end plate; median valve of female terminalia simple, slender and placed more terminal, apex touching the subapex of ovipositor; fifth instar immatures without tarsal arolium, instead with pulvilli on claws. These traits make the current systematic position of Togepsyllinae doubtful. Alternatively, Togepsyllinae share many similar characters with fossil pan-psyllids [extinct taxa included in Psyllomorpha by Bekker-Migdisova (1985), namely Pincombeidae, Protopsyllidiidae, Liadopsyllidae, Malmopsyllidae and Neopsylloididae] and whiteflies. With fossil pan-psyllids, the similarities include the half-modified jumping hind legs (compared with the none-jumping hind legs of Liadopsyllidae and Malmopsyllidae) (Ouvrard et al. 2010), the one-segmented aedeagus (particularly Syncoptozus, compared with Postopsyllidium) (Grimaldi 2003), frons fused with gena (compared with Postopsyllidium) (Grimaldi 2003). With whiteflies, the major similarities include the frons fused with gena, the pair of extra sclerites posterior to base of thoracic furcae (Fig. 61), presence of wax plates, one-segmented aedeagus, and absence of flag lobe on apex of valvula dorsalis of ovipositor. Similarities and differences among whiteflies, fossil pan-psyllids, Togepsyllinae and other modern psyllids, are listed in Table 2.
In the schematic phylogenetic tree (Burckhardt and Ouvrard 2012), Rhinocolinae, Spondyliaspidinae and Togepsyllinae were treated as sister groups; and Drohojowska (2015), using thoracic characters, also produced a phylogeny that assigned Togepsyllinae and Rhinocolinae as sister groups. The supportive characters of Drohojowska include: anapleural cleft hardly visible; ventral view of anterior protruding of katepiternum small and oval; meracanthus absent or as very small tubercle. In Rhinocolinae, the most similar with Togepsyllinae was Apsylla, a genus (monotypic) with completely unmodified metacoxa (Ouvrard and Burckhardt 2010). Judging from the SEM photograph given by Drohojowska (2015: Fig. 13), hind legs of Apsylla cistellata (Buckton) follow the same model with Togepsylla, being on a plane nearly parallel with that of the middle legs. While according to the illustration in Mathur (1975), A. cistellata possesses no flag lobe on the apex of valvula dorsalis.
In addition to the obvious synapomorphies of Aphalarinae members, i.e., mesothoracic trochantinal apodeme present on the anterior margin of the pleurite and metatibia with an open crown of apical spurs, Togepsyllinae and Rhinocolinae share other characters. They both have: a short clypeus; a pair of extra sclerites posterior to the base of thoracic furca; meracanthus absent or rather small; and the tubercle above the apical opening of metacoxa prominent. Most notable is the extra pair of sclerites posterior to the base of thoracic furcae. These shared characters may suggest a relatively close relationship between Togepsyllinae and Aphalaridae-Rhinocolinae.
Another species, Atmetocranium myersi (Ferris and Klyver) (Calophyidae: Atmetocraniinae), the sole member of the genus, is somewhat in resemblance with Togepsyllinae. Referring to a dry-mounted specimen and to the original description (Ferris and Klyver 1932: Fig. 15K), it was found that the species possesses the same type of hind legs as Togepsyllinae. This species also lacks the median suture on the vertex, flag lobes on valvulae dorsales of ovipositor, meracanthus on metacoxa, and metabasitarsal spurs. The immatures also possess two-segmented tarsi and lacks a tarsal arolium (Tuthill 1952). However, the male terminalia of A. myersi is of the common type, with proctiger and parameres oriented upward and the aedeagus two-segmented. Additionally, the gena is bisected as normal. Burckhardt and Ouvrard's (2012) assignment of  Atmetocranium into the higher Calophyidae was 'provisional', primarily based on the internal comb of apical spurs on metatibia and the one-segmented, asymmetrical antennal flagellum of immatures, but the family actually lacks defining synapomorphies (Burckhardt and Ouvrard 2012). Unfortunately, Atmetocranium is too scarce, and we did not have access to the slide-mounted specimens; thus, we do not know the details of its morphology, particularly those concerning the thorax. Therefore, currently, the relationship between Togepsyllinae and Atmetocranium remains uncertain. Phylogeny of Togepsyllinae seems unsolvable in the current situation, given the clear fact that there are only two known genera which are distinct from each other in many traits, indicating the possible existence of further extinct members of the group. On a greater scale, the current definition of Aphalaridae needs a phylogeny-based revision, to resolve its internal relationships and to test if Togepsyllinae is an independent taxon.