Capperia, Tutt, 1905

Alipanah, Helen, Gielis, Cees, Sari, Alireza, Sarafrazi, Alimorad & Manzari, Shahab, 2011, Phylogenetic relationships in the tribe Oxyptilini (Lepidoptera, Pterophoridae, Pterophorinae) based on morphological data of adults, Zoological Journal of the Linnean Society 163 (2), pp. 484-547 : 520-522

CAPPERIA COMPLEX

This complex comprises of all the known Capperia , Procapperia , Paracapperia , Intercapperia , and Eucapperia species. Here we consider Procapperia hackeri as a junior synonym of Capperia insomnis . The presence of a pair of hair tufts in the middle part of the specialized eighth sternite of males at the ventral surface is the only autapomorphy of C. insomnis . This species is closer to Procapperia species than Capperia in many characteristics of the male genitalia; nevertheless, it differs from all the known Procapperia species in the shape of the specialized eighth sternite of the male genitalia and the absence of a median pear-shaped structure in the seventh sternite of the female genitalia. Although host plants of Procapperia species and C. insomnis (here as a senior synonym of P. hackeri ) are members of Laminaceae, Procapperia species are highly specialized in their host plants and feed on different species of Scutellaria L. ( Nel, 1986; Matthews & Lott, 2005); whereas the known host plant for C. insomnis is Tinnea aethiopica Kotschy & Peyr ( Townsend, 1956) . It seems that C. insomnis is neither a member of Procapperia nor Paracapperia / Capperia . Molecular studies may help its exact position to be revealed. Therefore, here we consider it as a species incerta cedis.

In both the EW and SAW analyses, Paracapperia anatolicus and P. esuriens did not constitute a monophyletic clade. However, the monophyly of the genus Paracapperia was not rejected statistically (see Table 3). Oxyptilus anatolicus , was described by Caradja (1920), and transferred by Adamczewski (1951) to the genus Procapperia . Bigot & Picard (1986) transferred it to the subgenus Paracapperia (one of the three divisions of the genus Capperia ) and later, after they raised the subgenus Paracapperia to the genus level, it was considered to be a member of this genus. According to Bigot & Popescu-Gorj (1974) the characteristic feature of P. anatolicus is the special shape of the valva, which is narrow with an internal crest and a distal club. Gielis (1993) considered Paracapperia to be a sister group of Capperia , on the basis of two characters: strongly curved aedeagus with lateral processes and branching of vein Cu 2 in the fore wing from the lower angle of discal cell. The present study revealed that the latter homoplastic character was not common in all the Capperia species. In Capperia hellenica , vein Cu2 branches from beyond the lower angle of the discal cell (a homoplastic character), whereas in C. loranus , C. raptor , C. trichodactyla , and C. washbourni it originates from slightly behind the lower angle of discal cell. It is worth mentioning that P. anatolicus was the only member of this genus at that time and Gielis’s classification was based on this, single known, species. Until 2002, Paracapperia was a monotypic genus and distinguished from Procapperia and Capperia species by its twisted shape of the valva ( Arenberger, 2002). Gielis (2003) synonymized Trichoptilus inferens with Oxyptilus esuriens . Then he proposed a new combination for O. esuriens and moved it to Paracapperia .

On the basis of both the EW and SAW analyses, Paracapperia esuriens was the sister group of the clade consisting of all Capperia , Procapperia , Intercapperia , and Eucapperia species , and Paracapperia anatolicus ( Figs 16 , 17 ). The results of the present study revealed that P. esuriens differed from P. anatolicus in several aspects (e.g. characters 9, 25, 90, 120, 155, 159, and 160). Moreover, there are some other characters in all the members of the above-mentioned group except P. esuriens . For example, in P. esuriens the aedeagus is not S-shaped and the bulbus ejaculatorius arises from the phallobase nearly anteriorly, whereas in the remaining species (and probably Eucapperia bullifera ) the aedeagus is S-shaped and the bulbus ejaculatorius arises anterodorsally (homoplastic). Furthermore, the antrum in all of them is narrow, whereas it is wide in P. esuriens (homoplastic). The ductus bursae emerges nearly from the distal part of a cone-shaped folded area at the top of the bursa copulatrix (168: 1) in all species of this group but the character was scored as missing because of it not being clear in the genitalia of the examined female P. esuriens . If the same feature is present in this species, the character can be considered as a probable synapomorphy of the whole Capperia complex.

Paracapperia esuriens originally is an Afrotropical species, whereas P. anatolicus is distributed in the Palaearctic region ( Gielis, 2003). Although the rounded apex of each lobe of the bilobed specialized eighth sternite of males is a common feature of P. esuriens , P. anatolicus , Capperia raptor , and C. insomnis , this character did not support the monophyly of these species. Having said this, constraining the four species to be monophyletic was not rejected statistically (P = 0.8758 –1.000). Additionally, compared to the spoon-shaped valva of P. esuriens , the other three species are close to each other in their paddle-shaped valva. An almost similar idea was postulated by Adamczewski (1951). He stated that C. raptor and C. tamsi (which was later synonymized with Paracapperia anatolicus ) were close to each other in the shape of the valva, aedeagus, and eighth abdominal sternite of males. However, P. anatolicus , C. raptor , and C. insomnis , in spite of sharing some male genital characters, did not constitute a monophyletic clade either. It seems that P. esuriens belongs to a different genus and here it is considered as species incerta cedis.

All the Procapperia species except P. hackeri constituted a monophyletic clade. In all of the consensus trees obtained from the EW and SAW analyses, this monophyletic clade is the sister group of a major clade consisting of Intercapperia , Eucapperia , all the Capperia species , and Paracapperia anatolicus (clade Y, Fig. 16 ). In all the Procapperia species except P. hackeri , which here is considered as a junior synonym of Capperia insomnis , the specialized eighth sternite of males has an internal flap. This homoplastic character is also present in Intercapperia scindia .

A close affinity of Procapperia and Capperia was proposed by Bigot (1966) based on the shape of the ninth tergum (tegumen). A sister-group relationship between Procapperia and the two genera Paracapperia and Capperia was postulated by Gielis (1993). Although the shape of the ninth sternite (the specialized eighth sternite in the present study) of males was used by Gielis (1996) to separate Procapperia from Paracapperia , choosing the same key characters may create some difficulties in their separation. According to Gielis (1996), in all the Procapperia species , the apex of the specialized eighth sternite of males is bifurcated, whereas in Paracapperia species it is short, blunt, and covered with hair brushes. However, in all available materials examined in the current study, the distal part of the specialized eighth sternite of males, even in Paracapperia species , was bifurcated. Most of the Procapperia species at the distal part of their specialized eighth sternite of males have a pair of lateral projections that may give a furcating appearance, but these cannot be considered as real furcae.

Arenberger (2002) separated Palaearctic species of Procapperia from Capperia based on two characters: (1) in Procapperia , the seventh sternite of the female genitalia, which covers the ostium, is pear-shaped, whereas in Capperia it is not pear-shaped; (2) in the male genitalia of Procapperia species , the valva has no prominent basal process (the costal process in the present study), whereas in Capperia the process is present. However, these characters are not present in all Palaearctic members of Capperia and Procapperia . For example, as stated by Zagulajev (1986) and Gibeaux (1997), and as we found in the present study, the costal process of the valva is present and prominent in Procapperia kuldschaensis . In Procapperia amira , P. maculatus , and P. orientalis , a basal swelling is present near the base of the valva in the costal region. Moreover, the presence of a median pearshaped structure in the seventh sternite of the female genitalia is not common in all known Procapperia species. Based on the present study, this structure is absent in all examined specimens of Procapperia kuldschaensis (collected in Tadzhikistan). It is also absent in Procapperia hackeri , giving an additional piece of evidence to exclude P. hackeri from Procapperia . Having said this, excluding the latter character from the analysis had no influence on the resulting trees.

Zagulajev (2002) described a new Procapperia species , namely Procapperia tadzhica . No specimen was available to study, but according to the original description, it is closer to Capperia species than Procapperia spp. This species is close to Capperia salanga in the shape of the valva and aedeagus, and like C. salanga , its specialized eighth sternite has a pair of hairy flaps near the middle part. However, the shape of the costal process of the valva seems to be different. For these reasons, P. tadzhica is only here considered as a member of Capperia (see Nomenclatural changes); however, examination of the holotype may reveal that P. tadzhica and C. salanga are the same species.

Even though there was no synapomorphy to support the monophyly of the Eucapperia – Intercapperia group, in all the consensus trees obtained from the EW and SAW analyses, these two genera constituted a monophyletic clade that was a sister group of all Capperia species , except C. raptor . Platyptilia bullifera has recently been moved to Eucapperia by Gielis (2009) and considered as a senior synonym of Eucapperia continentalis . Although Eucapperia differs from Intercapperia and the remaining genera of clade AE ( Fig. 16 ) in the wing shape and pattern, it is close to the members of the latter clade in male genitalia characters and is still classified in this group. This result confirms the importance of genital characters in generic classifications.

In all the consensus trees obtained from both the EW and SAW analyses, Capperia raptor and C. insomnis were not gathered in the same clade with other Capperia species. The former species was classified in Capperia by Barnes & Lindsey (1921) according to the shape of its male genitalia. However, as recovered here, it is the sister group of a clade including Intercapperia , Eucapperia , and the remaining Capperia species except C. insomnis . There is more evidence that may support the exclusion of C. raptor from all known Capperia species. For example, C. raptor differs from other Capperia species in some characteristics of the specialized eighth sternite of males (characters 74, 76, 78, 79, and 81), shape of the gnathos (character 99), anellus (character 115), and shape of the valva and its processes (characters 120, 127, and 142). In addition, characters 91, 99, 115, and 142 are autapomorphic characters for this species. Capperia raptor also differs from all Capperia species in feeding behaviour. Whereas all known Capperia species feed on Laminaceae, the only known foodplant for C. raptor is Geranium caespitosum James (Geraniaceae) . It is worth mentioning that some other host plant species from Asteraceae and Plantaginaceae have been recorded for C. ningoris , C. trichodactyla , and C. maratonica ( Matthews & Lott, 2005) . Considering all these, C. raptor may be the only representative of a new genus (will be described elsewhere).

According to Gielis (1993), pronounced sclerotized ridges and the hairy tip of the valva may be considered as autapomorphies of Capperia . However, we mention here some interspecific variations in the shape of the valva amongst the species of this genus. In all species, except C. raptor , the valva is spoonshaped. In Capperia salanga , the tip of the valva is flat but in other species of the genus it is cube-shaped. The arrangement of the hairs at the tip of the valva is also different. Furthermore, the costal process of the valva in most species is straight (homoplastic character), whereas in C. salanga it is slightly curved (in C. raptor the costal process is absent). Capperia salanga is the only Capperia species that has a pair of hairy flaps in the middle part of the specialized eighth sternite at the ventral surface. We also found that in C. hellenica , the median sclerotized projection at the posterior margin of sternite II is absent in both sexes, whereas in all other Capperia species it is present only in males (a homoplastic character). It seems probable that the low values of tree confidence for the clade including all Capperia species (except C. raptor and C. insomnis ; Fig. 16 ) are the result of these interspecific variations within the genus.

Figures

Figure 16. Strict consensus trees of six equally most parsimonious trees obtained in the initial analysis using equally weighed characters. Numbers below the branches show Bremer support values and the capital letters show the clades included in Oxyptilini and are discussed in the text. Tree length = 943, consistency index = 0.4549, retention index = 0.7976.

Figure 17. Strict consensus trees of six equally most parsimonious trees obtained in the equally weighted analysis (left), and that of 380 most parsimonious trees obtained from successive approximation weighting based on maximum retention index values (right). Numbers above and below (underlined) the branches show bootstrap and jackknife values, respectively.