Four new species of Aspidiotini (Hemiptera, Diaspididae, Aspidiotinae) from Panama, with a key to Panamanian species

Abstract Four new species of armored scale insect, Clavaspisselvaticasp. nov., Clavaspisvirolaesp. nov., Davidsonaspistovomitaesp. nov., and Rungaspisneotropicalissp. nov., are described and illustrated from Panama. We also transfer two previously described species of Panamanian Aspidiotini to new genera, Hemiberlesiacrescentiae (Ferris) comb. nov. and Rungaspisrigida (Ferris) comb. nov., and report the first record of Selenaspidopsisbrowni Nakahara in Panama. A key to the species of Aspidiotini occurring in Panama is provided.


Introduction
Armored scales are the most species-rich family of scale insects, comprising over 2600 species in 418 genera (García Morales et al. 2016). The family is characterized by the complete loss of legs and reduction of antennae in adult females, fusion of the posterior abdominal segments into a pygidium, and the formation of a waxy test (Takagi 1990). Like all members of the suborder Sternorrhyncha, armored scales are strictly phytophagous. Many species of armored scales are pests of agricultural commodities (Miller and Davidson 2005). Heavy infestations inhibit photosynthesis through chlorophyll depletion and crowding of leaf surfaces, reducing plant vigor. Additionally, visible infestations and damage reduce the value of produce and nursery stock (Kosztarab 1990;Miller and Davidson 2005). However, unlike most scale insects, armored scales do not contribute to the growth of sooty molds on hosts because they do not produce honeydew (Henderson 2011).
Some armored scale insect species are extremely polyphagous, with host ranges among the widest known for any herbivorous insect, comprising in some cases over 100 families of plants (Normark and Johnson 2011;Ross et al. 2013;García Morales et al. 2016). The most highly polyphagous species have a strong tendency to be economic pests (Normark and Johnson 2011;Ross et al. 2013;Normark et al. 2014). Because armored scale insects appear to have essentially random dispersal via windblown larvae, Hardy et al. (2015) hypothesized that their host ranges are likely to reflect the plant diversity of their habitats, and that extreme polyphagy may have evolved in habitats with extreme plant diversity, such as tropical rainforests. Because of the economic importance of armored scale insects, they have been extensively sampled on cultivated plants, especially orchard crops and ornamentals (Rosen 1990;Miller and Davidson 2005). But their diversity, abundance, and host associations in natural environments are poorly known, and this is particularly true for tropical rainforests. Since 2010, one of us (BBN) has been systematically sampling armored scale insects in tropical forests. Two of the goals of this effort are to test for cryptic diversity within apparently polyphagous species and to test whether a species' local abundance is correlated with its host range. Results of tests of these hypotheses using samples from Panama and Borneo are reported in Peterson et al. (2020). Briefly, cryptic diversity is found within some apparently polyphagous species within their native ranges, but some invasive species are truly polyphagous. And local abundance is positively correlated with host range. Another goal of the rainforest sampling effort is to discover and describe new species of armored scale insects, which is the purpose of this article. Specifically, here we describe four new species within the tribe Aspidiotini collected from Panama.
Armored scales are currently classified into four subfamilies: Ancepaspidinae, Aspidiotinae, Diaspidinae, and Furcaspidinae . Aspidiotini is a large tribe within subfamily Aspidiotinae that includes many pest species that are globally invasive and economically damaging (Schneider et al. 2018). To date, 54 species of Aspidiotini in 16 genera have been recorded from Panama (García Morales et al. 2016;last accessed 31.iii.2021). In addition to the descriptions of four new species, this article includes the first report of Selenaspidopsis browni Nakahara from Panama. Additionally, this article assigns two Neotropical species to the genus Rungaspis, whose species are otherwise restricted to Africa and the southwestern Palearctic. With these records included, 58 species from 18 genera in Aspidiotini are known to occur in Panama, comprising roughly half of the total armored scale fauna for this country (58 out of 118 species reported in ScaleNet) (García Morales et al. 2016). The majority of these species are likely native to the Neotropics (Ferris 1941(Ferris , 1942Deitz and Davidson 1986), but many are broadly distributed and are considered major, minor, or potential pests (Miller and Davidson 1990;Schneider et al. 2019). Species that are non-native to this region include members of Aspidiella, Aspidiotus, Chrysomphalus, and Selenaspidus, which are widespread pests likely originating from the Australasian, Oriental, and Afrotropical regions (Schneider et al. 2018). An identification key to the species of Aspidiotini found in Panama is provided.

Material and methods
The sampling locality for new species described in this paper was the canopy crane in San Lorenzo National Park, Colón (9.2802°N, 79.9754°W). The locality was chosen because it offered access to the canopy via the crane and because every tree was reliably identified to species. The first survey was conducted in June 2012 by Geoffrey E. Morse and BBN, and the second in January 2015 by G. E. Morse, Daniel A. Peterson, Hannah Shapiro, and Shannon Trujillo. A full description of the sampling protocol is given in Peterson et al. (2020). Briefly, in each survey, investigators sampled all the tree species accessible from the canopy crane, and sampled multiple individuals of the more abundant species. Foliage of each sampled tree was searched visually for 20 person-minutes. Leaves that appeared to be infested with armored scale insects were collected into plastic bags, along with a 20 cm twig sample and 20 cm 2 bark sample. Collected material was refrigerated and examined under a dissecting microscope within 5 days; live armored scale insects were transferred to 100% ethanol. Subsequently, sampled scale insects were subjected to a joint morphological / molecular sample preparation that resulted in a sample of purified genomic DNA and a permanent microscope slide mount of the specimen's cuticle, following the method described in Normark et al. (2019).
In this paper, morphological terminology conforms to descriptions and illustrations provided by Schneider et al. (2019) and Miller and Davidson (2005). Vouchering of specimens was completed following the protocols described by Normark et al. (2019). Measurements were made on a Zeiss Axio Imager.M2 (Carl Zeiss Microscopy, LLC, White Plains, NY, USA) microscope with the aid of an AxioCam and AxioVision software. Illustrations were made using a Nikon Optiphot compound microscope (Nikon USA, Melville, NY, USA) with the aid of a camera lucida. Slide-mounted specimens were examined by the authors under phase contrast and DIC microscopy. The abbreviations L1, L2 and L3 refer to the median, second, and third pygidial lobes, respectively.
Depositories are abbreviated as follows: Without disc pores associated with anterior or posterior spiracles. Lobes. L1 well developed, slightly wider than long, inner margins near parallel, with 1 notch on each side or without notches, rounded apically; space between lobes approximately 0.25 times width of L1. L2 and L3 absent. Plates cylindrical, narrow, pointed at apex, simple or with a few fine tines, about as long as L1; 2 plates present in first space, often with 1 or 2 tines near apex giving bifurcate or trifurcate appearance; 1 or 2 plates present in second space, simple or with minute tines; plates absent between L1. Ducts. Dorsal macroducts of 1-barred type, with 2-3 macroducts arising from first space, 8-10 arising from second space, and 7-8 arising from third space in singular rows. Series of marginal macroducts with wide orifices extending from mesothorax to abdominal segment II; at least two present per segment. Groups of ventral submarginal microducts occurring on head, thorax, and abdominal segments I-V. Paraphyses. With 1 pair of paraphysis-like basal scleroses near mesal margins of L1; 1 pair of paraphyses in first space, paraphysis arising from lateral margin of L1 slightly longer than paraphysis arising from medial margin of L2, both mushroom-like in shape with distinctive dome or cap at anterior end; 1 pair of small clavate paraphyses in second space. Anal opening longer than wide, 11-14 μm long, 5-7 μm wide, positioned 17-25 μm (1.5-2 anal lengths) from the base of L1, located within posterior third of pygidium. Perivulvar pores few, 2-6 pores in total, divided into 2-4 groups, with 1-4 in each group.

MIUP
Remarks. This new species is most similar in appearance to C. coursetiae (Marlatt) with subtle differences distinguishing the two. Submarginal groups of microducts form a semicircle around the head, thorax, and pre-pygidial abdominal segments of C. selvatica but are more diffusely scattered in C. coursetiae, not organized in an obvious semicircular ring. In C. selvatica, at least two large macroducts are present on the mesothorax, while in C. coursetiae only one at most is present, falling near the posterior margin of the mesothorax. The plates are nearly as long as L1 and fringed in C. selvatica but are short and simple in C. coursetiae. This species is also similar to C. subsimilis (Cockerell) in body shape and the shape of L1 but can be distinguished by possessing perivulvar pores on the pygidium (absent in C. subsimilis).
Host plant. Apeiba aspera Aubl. (family Malvaceae). Etymology. The epithet selvatica is the Latin adjective meaning wild, literally "of the forest" (selva). Our choice of this name is influenced by the fact that in modern Spanish, the word selva is identical to its Latin ancestor in form, but now refers specifically to tropical rainforest.
Distribution. Panama (Colón). Body outline oval, nearly circular in smaller individuals (< 600 μm long), becoming elongate-oval in larger individuals. Derm membranous throughout at maturity except for pygidium. Antennae simple, each with one long seta. Distance between antennae 40-100 μm. Without disc pores associated with anterior or posterior spiracles. Lobes. Pygidium with 2 pairs of lobes; L1 well developed, separated by space about one-fifth width of L1, lobes slightly wider than long, inner margins near parallel, with 1 medial and 1 lateral notch, rounded apically; L2 forming sclerotized point, about onequarter to one-third size of L1, with 1 lateral notch; L3 absent, indicated at most by small, lightly sclerotized projection of pygidial margin. Plates. All plates simple; with or without fine plates in slight space between L1; with 2 pointed plates in first space; plates absent in second space; five simple microduct-bearing plates present laterad of L3, nearly as long as L1. Ducts. Dorsal macroducts of 1-barred type, slender, with orifices narrower in diameter than ventral microducts, restricted primarily to margin with one submarginal duct anterior to seta marking segment VI; 1 between L1, with 3-4 marginal ducts in first space, 2 marginal ducts in second space; with few short macroducts occurring on submarginal areas of pre-pygidial segments. Ventral microducts slightly wider in diameter than dorsal macroducts and present in small submarginal groups on pre-pygidial abdominal segments and segment V. Paraphyses. L1 each with a paraphysis-like basal sclerosis toward medial margin, slightly smaller than lobe; in first space, 1 clavate paraphysis arising from lateral angle of L1, 1 arising from me- sal angle of L2, posterior-most paraphysis slightly longer than L1; 2 smaller clavate paraphyses arising from mesal margin of L3. Anal opening oval, 8-13 μm in length, 4-6 μm in width, positioned 20-23 μm from base of L1, located within posterior third of pygidium. Perivulvar pores absent.

Clavaspis virolae
Remarks. This species is placed in the genus Clavaspis MacGillivray on the basis of the robust clavate paraphyses, small anal opening, and basal sclerosis of L1, resembling that of Clavaspis ulmi (Johnson). The paraphyses are not as elaborately developed as those of most Clavaspis species, but they are more developed than some species that have recently been recognized as members of Clavaspis on the basis of molecular phylogenetics -C. perseae (Davidson) and C. patagonensis Schneider, Claps, Wei, Normark & Normark (Normark et al. 2014;Schneider et al. 2020). Clavaspis virolae is similar to Clavaspis ulmi, but differs in having L2 present, plates fewer, dorsal macroducts fewer, medial paraphysis of first space less developed, and ventral macroduct orifices larger than those of dorsal macroducts. Clavaspis virolae also resembles species of Hemiberlesia Cockerell, especially H. ignobilis Ferris and H. ocellata Takagi & Yamamoto, but differs in having a smaller anal opening and fewer plates. It further differs from H. ignobilis in having L2 present and ventral macroduct orifices larger than those of dorsal macroducts, and from H. ocellata in having 2 pairs of conspicuous paraphyses present, L3 absent, and notching of L1 and L2 less deep. Yet another genus that C. virolae resembles is Diaspidiotus Berlese: the axes of L1 and L2 seem to converge slightly, causing the species to key out as Quadraspidiotus MacGillivray, now a synonym of Diaspidiotus, in Ferris's (1942) key. But this is not as good a fit, as Diaspidiotus species lack basal scleroses of L1. It is also biogeographically less plausible, as Diaspidiotus is overwhelmingly a temperate Holarctic group. There exist Neotropical species assigned to Diaspidiotus, but these may be misplaced. The only such species reported from Panama, D. crescentiae Ferris, has a large anal opening and basal scleroses of L1, and is best regarded as Hemiberlesia crescentiae (Ferris)  Description (N = 2). Adult female not pupillarial. Appearance in life not recorded. Slide-mounted adult female 870-1060 μm long, 670-790 μm wide, broadest at mesothorax. Body outline broadly obovate. Antennae simple, each with one conspicuous long seta. Distance between antennae 160 μm. Without any disc pores associated with anterior or posterior spiracles. Lobes. Pygidium with 3 pairs of lobes extending out from posterior margin, well sclerotized. L1 large, apically convergent, each lobe oval in shape, with minute notch near midpoint of outer margin; L2 distinctly smaller than L1, with 1 or 2 small notches on outer margin; L3 similar in size and shape to L2, with 1 or 2 notches on outer margin and 0 or 1 notch on inner margin. Plates. Without plates between L1; with 2 plates between L1 and L2, flabellate, apically fringed, each slightly longer than L1, much longer than L2; three between L2 and L3, flabellate, apically fringed, each plate longer than L3; three anterior to L3, branched and elaborately fringed on apical and lateral margins, much longer than L3, each with internal microduct. Ducts. Dorsal macroducts of 1-barred type, dorsal submarginal macroducts about same size as marginal macroducts, long (120-140 μm) and narrowly ribbonlike, with minute orifices, few, only 10-15 on each side of pygidium. Also, with faux duct orifice on dorsum immediately anteriad of L1 -circular structure slightly larger than duct orifices, but without duct. Pre-pygidial dorsal macroducts few, shorter than those on pygidium, confined to margin and submargin, absent on segments III and IV, two present on each side of segments II, I, metathorax, and mesothorax. Ventral microducts shorter and thinner than dorsal macroducts, with a few present on submargin of each segment from abdominal segment V to prothorax. Paraphyses. Three pairs of paraphyses present on each side of pygidium, variable, with lateral member of each pair often minute or absent. Medial pair of paraphyses anteriad of L1, medial member of pair arising from near inner angle of L1, extending nearly to anus and terminating in rounded knob, lateral member of pair minute, forming part of sclerotized rim of faux duct orifice; pair of paraphyses between L1 and L2 also with medial paraphysis much larger than lateral paraphysis; pair between L2 and L3 usually about equal to each other in length, lateral member of pair sometimes obsolete. Anal opening nearly circular, maximum diameter 8 μm, located 23-25 μm (about 3 times diameter) from base of L1. Perivulvar pores absent.
Remarks. The only other known species in this genus is Davidsonaspis aguacatae (Evans, Watson, and Miller), found on avocados in Mexico. D. aguacatae had originally been assigned to Abgrallaspis Balachowsky (Evans et al. 2009), but was later reassigned to a new genus Davidsonaspis Normark (Normark et al. 2014). The new species can be distinguished from D. aguacatae in having a series of 3 plates anterior to L3, each as broad as L3 and elaborately fringed on apical and lateral margins; in D. aguacatae, plates anterior to L3 are narrower than L3 and only slightly fringed. D. tovomitae otherwise closely resembles D. aguacatae, and the two species form a clade in published molecular phylogenetic trees (Schneider et al. 2018;Normark et al. 2019). The structure we refer to as a faux duct orifice anteriad of L1 is illustrated by Evans et al. (2009) but not mentioned in their description. In one of their 2 illustrations of the pygidium of D. aguacatae the structure is shown with a central dot, as if it were the circular base of a seta, but in D. tovomitae no seta is present there.
Host plant. Description. Adult female (N = 10) in some cases pupillarial, enclosed within sclerotized cuticle of 2 nd instar; some individuals non-pupillarial. Appearance in life not recorded. Slide-mounted adult female 350-610 μm long (holotype 540 μm, median 540 μm), 280-500 μm wide (holotype 410 μm, median 420 μm), broadest at mesothorax. Body outline broadly oval, with slight indentation between prothorax and mesothorax. Derm membranous throughout at maturity in pupllarial individuals; cephalothorax and pygidium becoming sclerotized at maturity in some non-pupillarial individuals. Antennae simple, each with one long seta. Distance between antennae 51-73 μm. Eye a submarginal dorsal tubercle on prothorax. Without disc pores associated with anterior or posterior spiracles. Venter of mesothorax with about 6 transverse, irregular rows of sclerotized spicules in submedial area, posterolaterad of mouthparts. Lobes. Pygidium with 1 or 2 pairs of lobes; L1 well developed, subquadrate, with parallel inner margins separated by exceedingly narrow space, lobes slightly longer than wide, rounded apically, with 1 large notch near apex on lateral margin and 0-1 notch near apex on medial margin; L1 each with well-developed basal sclerosis, slightly narrower and longer than lobe; L2, when fully developed, forming a small, sclerotized projection, about one-third length of L1 and much narrower, without notches; L2 often absent or represented by a membranous projection or low, sclerotized point; L3 absent. Plates. With 2 narrow, elongate plates in first space, slightly fringed, with a few tines,  and 1 or 2 simple plates laterad of position of L2; no other plates present. Ducts. Dorsal macroducts of 1-barred type, slender, much broader than ventral microducts, few in number, restricted to margin of pygidium; with 1-3 (usually 2) ducts in first space, 0-2 (usually 1) immediately laterad of L2, and 0-1 (usually 0) laterad of seta marking segment VI, making a total of only 1-4 ducts (usually 4) on each side of pygidium. Ventral microducts exceedingly narrow, present along pygidial margin and scattered in submedial areas of other segments. Paraphyses absent. Anal opening subcircular, 8-11 μm in length and width, positioned 17-37 μm from base of L1, located within posterior half of pygidium. Perivulvar pores absent.
Second-instar female (N = 8) Appearance in life not recorded. Slide-mounted second-instar female 340-620 μm long (median 460 μm), 270-400 μm wide (median 340 μm), broadest at mesothorax. Body outline oval. Antennae simple, each with one long seta. Distance between antennae 54-96 μm. Without disc pores associated with anterior or posterior spiracles. Lobes. Pygidium with 3 pairs of well-developed lobes; L1 subquadrate, with parallel inner margins separated by exceedingly narrow space, lobes slightly longer than wide, rounded apically, with 1 large notch near apex on lateral margin and 0-1 notch near apex on medial margin; L1 each with well-developed basal sclerosis, slightly narrower and longer than lobe; L2 nearly as long as L1 but much narrower, rounded at apex, without notches or with slight notch on lateral margin; L3 subtriangular, slightly narrower and shorter than L2, without notches. Plates. Without plates between L1. With 2 narrow plates in first space, 2 broader plates in second space, and a series of 5 or 6 plates laterad of L3. All plates similar in length to adjacent lobes and fringed at apex, with plates anterior to L3 becoming progressively lower and less fringed anteriorly. Plates of the first and second spaces subtended by conspicuous ducts, about a third as wide as dorsal macroducts and nearly as long, much wider and longer than ventral microducts. Ducts. Dorsal macroducts of 1-barred type, broad, all submarginal; with 2 ducts in a short row arising from first space, 2 in the second space, and 1 laterad of L3, making a total of 5 on each side of pygidium. Ventral microducts exceedingly narrow, short, present along pygidial margin and scattered in submedial areas of other segments. Paraphyses absent. Anal opening oval to subcircular, Informal synonyms. Specimens of R. neotropicalis have appeared in published analyses and phylogenetic trees, where they were labeled "UG3995 ud3995" (Schneider et al. 2018;Normark et al. 2019), "UG3953 ud3953" (Schneider et al. 2018), or "Rungaspis ud3995" (Peterson et al. 2020).
Remarks. This is an unusual species both in its life history, showing intraspecific variation in the pupillarial habit, and in its biogeography, having affinities to African species. Some slide-mounted specimens are unequivocally pupillarial, having well-developed 1 st instars inside of adult females that are themselves inside of 2 nd -instar cuticles. More often than not, these adult females are flipped inside their puparia, with their head at the posterior end of the puparium. Other specimens are apparently nonpupillarial, and some of these have a sclerotized cephalothorax, a feature not seen, to our knowledge, in adult females of any pupillarial species. We had originally intended to describe the pupillarial and non-pupillarial forms as two different species, but the three sequenced gene regions show no differences between them and there are no consistent morphological differences either; therefore, we consider them to comprise a single species that includes both pupillarial and non-pupillarial developmental phenotypes. The second instar has a more completely developed secretory system than the adult, with more ducts, plates, and lobes -a pattern typical of pupillarial species and opposite to what is typical of non-pupillarial species. This may imply that this species is derived from a pupillarial ancestor and that the non-pupillarial form represents a secondary loss of the pupillarial habit.
Molecular phylogenetic studies have shown that R. neotropicalis has affinities with African species. Probably the best analysis is a recent study of Aspidiotini (Schneider et al. 2018), which shows R. neotropicalis nested within a clade of African Aspidiotus species (A. fularum Balachowsky, A. elaeidis Marchal, and an undescribed species from Uganda), with R. neotropicalis sister to A. fularum. R. neotropicalis was also included in a broader study of Diaspididae , where it appears in a clade that consists mostly of African species (A. elaeidis, Selenaspidus kamerunicus Lindinger, S. articulatus Morgan, Dynaspidiotus rhodesiensis (Hall), and Entaspidiotus lounsburyi (Marlatt)) but that also includes one other New World species (Rugaspidiotus arizonicus (Cockerell)). It is possible that R. neotropicalis is an African species that is invasive in the Neotropics, similar to Selenaspidus articulatus, which is the single most abundant diaspidid species at the site where R. neotropicalis was collected (Peterson et al. 2020). But if this species is from Africa, it does not seem to have ever been found there. Based on the available evidence we regard it as a native Neotropical species, perhaps one resulting from an ancient trans-Atlantic dispersal event.
We tentatively place this species in the genus Rungaspis Balachowsky. Rungaspis presently comprises four species distributed in Africa and the southwestern Palearctic. Rungaspis neotropicalis resembles the other species of Rungaspis in having large basal scleroses of L1, reduced L2 and L3, cephalothoracic sclerotization at maturity (in non-pupillarial specimens), dorsal ducts with sclerotized orifices, and simplified plates located only in the first and second interlobular spaces. African Rungaspis species differ from R. neotropicalis in having conical plates without fringes (vs. slightly fringed) and numerous narrow dorsal ducts (vs. few broad ducts). We considered three other possible placements for the species. One was the genus Aspidiotus Bouché. Rungaspis neotropicalis resembles Aspidiotus species in having basal scleroses of L1 and fringed plates, and molecular evidence indicates that its closest known relative is an African species of Aspidiotus. But we concluded that R. neotropicalis shares a greater number of characters with Rungaspis. Furthermore, Aspidiotus is radically non-monophyletic, and the mostly African clade to which R. neotropicalis belongs should probably be recognized as a distinct genus anyway (Schneider et al. 2018). Another possible placement we considered was the genus Helaspis McKenzie. Helaspis is a New World genus that "appears to suggest Aspidiotus more strongly than any known genus" (McKenzie 1963). With R. neotropicalis it shares basal scleroses of L1 and a sclerotized cephalothorax. But Helaspis has other extraordinary features -conical plates and bilobed L3that seem to indicate an affinity with the tribe Gymnaspidini rather than Aspidiotini . Rungaspis neotropicalis lacks these characters and is clearly a member of Aspidiotini. We also considered erecting a new genus for R. neotropicalis -this is the course taken by many diaspidid systematists faced with such an unusual species -but we concluded that that was not appropriate in this case given the evidence for affinity with Rungaspis.
Morphologically, R. neotropicalis also closely resembles Aspidiotus rhusae (Brain), a pupillarial species known from South Africa. The two species share a similar overall body shape, L1 with basal scleroses, absence of L3, absence of perivulvar pores, and presence of just a few slightly fringed plates and just a few broad, one-barred dorsal ducts near the pygidial margin. Characters that distinguish R. neotropicalis from A. rhusae are as follows (character of A. rhusae given in parentheses): L2 much narrower than L1 or absent (L2 nearly as broad as L1); space between L1 exceedingly narrow, without plates (space between L1 with pair of apically fringed plates); 4 or fewer dorsal ducts present on each side of pygidium (5 or more ducts present); 1-3 microducts present near each posterior spiracle (cluster of 5 or more ducts in this position); transverse rows of minute spicules present on mesothorax posterolaterad of mouthparts (absent); body margin slightly indented between prothorax and mesothorax (entire); eye a submarginal dorsal tubercle (eye marginal). The Neotropical species that R. neotropicalis most closely resembles is Aspidiella rigida Ferris. The two species both have L1 with basal scleroses and closely approximated medial margins, other lobes reduced or absent, cephalothorax becoming sclerotized at full maturity, and perivulvar pores absent. Characters that distinguish Rungaspis neotropicalis from Aspidiella rigida are as follows (character of A. rigida given in parentheses): plates present (absent); dorsal ducts of pygidium broad, much broader than ventral microducts, confined to margin and submargin (narrow, similar to ventral microducts, widely scattered); anus in posterior half of pygidium (anterior half ).
Our study of Neotropical and African species that resemble Rungaspis neotropicalis has further led us to conclude that Aspidiella rigida belongs in the genus Rungaspis, and we transfer it accordingly: Rungaspis rigida (Ferris) comb. nov. Ferris (1941) remarked, "It is with much doubt that this species is here referred to the genus Aspidiella. In its pygidial characters it resembles the type genus closely enough except for the entire absence of plates and the absence of the perivulvar pores... In the heavy sclerotization of the entire body it is peculiar and distinctive." In each of these characters it resembles Rungaspis species more than Aspidiella species. Ferris further expressed puzzlement that an Oriental and Australian genus such as Aspidiella would include a species that was apparently native to the Neotropics. A biogeographic connection between the Neotropics and Afrotropics is better documented (by Rungaspis neotropicalis and in groups such as Diaspis Bouché) and less of a surprise.
Host plant. Marila laxiflora Rusby (family Calophyllaceae) Etymology. The specific epithet is a Latin adjective; here it alludes to this species' unusual biogeography as a Neotropical member of a mostly African clade.

Key to species of Aspidiotini from Panama based on adult females
The key incorporates some modified excerpts drawn from Ferris (1942), Deitz and Davidson (1986)