Phoebe Audinet-Serville, 1835

Santos-Silva, Antonio, Heffern, Daniel, Botero, Juan Pablo, de, Francisco Eriberto & Nascimento, L., 2020, Notes, new records, new combinations, a new genus and three new species in Hemilophini (Coleoptera: Cerambycidae: Lamiinae), Insecta Mundi 2020 (785), pp. 1-25 : 7-8

On Phoebe Audinet-Serville, 1835

Phoebe is involved in a series of mistakes since its original description (and even before), which includes the type species of the genus, and its characterization and limits. Herein we try to solve these problems and propose a new synonymy.

Lepeletier and Audinet-Serville (1825) described Saperda phoebe and included it in their “2 e Division. Corps alongé”, and in the “2 e Subdivion. Antennes de onze articles dans les deux sexes.” Audinet- Serville (1835) described Agapanthia and affirmed that the antennae are 12-segmented. Agapanthia was divided into two subgenera, A. ( Agapanthia ) and A. ( Phoebe ). The latter included three species, A. octomaculata Audinet-Serville, 1835 ; A. cornuta (Olivier, 1800) ; and A. bicornis (Olivier, 1800) (currently, all of them in Phoebe ). Audinet-Serville (1835) affirmed that A. octomaculata is Saperda phoebe : “ Agapanthia octomaculata . – Saperda Phoebe, Encycl. méthod., tom. X, pag. 335, nº 2. Du Brésil.” Apparently, Audinet-Serville (1835) was renaming the species because he was using Phoebe as a subgeneric name. Accordingly, the information regarding the number of antennal segments in Phoebe is contradictory: 11 in Lepeletier and Audinet-Serville (1825), but 12 in Audinet-Serville (1835). However, although Olivier (1800) did not report the number of antennal segments in Saperda bicornis (we do not know if the holotype survived and, if so, where it is, original source was given as collection Raye), the specimens currently assigned to this species agree well with the drawing provided by Olivier (1800), and the antennae are 11-segmented. He also did not record the number of antennal segments in Saperda cornuta (we do not know if the holotype still exists and, if so, where it is, original source was given as collection Gevers). Unfortunately, the specimens currently assigned to this species do not agree well with the original description, and it is thus not possible to be sure if the antennae in the true S. cornuta are 11 or 12-segmented. However, they are 11-segmented in S. cornuta sensu auctorum. Saperda phoebe and Agapanthia (Phoebe) octomaculata will be discussed below.

Laporte (1840) considered Phoebe as a genus different from Agapanthia . Pascoe (1858) described Phoebe cretifera from Brazil, but did not mention the number of antennal segments.However, photographs of the holotype show 12 segments. Desmarest (1860: 328) designated Saperda bicornis Olivier, 1800 as the type species of Phoebe : “Un dernier genre, tantôt réuni aux Saperda , et tantôt aux Agapanthia , est celui des Phoebe, Serv. , caractérisé par sa tête portant en avant une lunule sailante, dont les pointes s’élévent plus ou moins en manière de corne, à face antèrierure courte et à front bombé, et qui a un corps peu svelte: espèce typique, S. bicornis, Oliv., de Cayenne. ” Desmarest (1860) used both, “espèce typique” and “type.” However, even when he used “espèce typique”, he meant the type species and not a “typical species of the genus” ( ICZN 1999: Article 67.5.1). This is very clear from some examples where he used “genre typique” in the same work, e.g. “Mégalopites… MEGALOPUS, Fabr., genre typique...”. Thomson (1864) designated Agapanthia (Phoebe) octomaculata as the type species. Evidently, the designation by Thomson (1864) is not valid.

Bates (1866) transferred Phoebe bicornis to Amphionycha Dejean, 1835 (= Adesmus Audinet-Serville, 1824 ). Gemminger (1873) synonymized P. cretifera with P. octomaculata . Aurivillius (1923) formalized the synonymy between Saperda phoebe and Agapanthia (Phoebe) octomaculata .

According to Martins and Galileo (1992) (translated): “There are relatively few genera in this tribe [ Hemilophini ] whose species have antennae with twelve articles, all so far described by Lane (1956, 1966).” They listed Murupeaca Martins and Galileo, 1992 , Phoebella Lane, 1966 , Gagarinia Lane, 1956 , Tabatinga Lane, 1966 , Purusia Lane, 1956 , and Juninia Lane, 1966 . Phoebe was not mentioned, although the original description made clear that it has antennae 12-segmented (partially incorrect information), and the species wrongly mentioned as being the type species has antennae 12-segmented. Later, Martins and Galileo (1993) described one more genus in Hemilophini with antennae 12-segmented: Sibapipunga .

Martins and Galileo (1998, 2014b) redescribed Phoebe phoebe , but did not mention the number of antennal segments. However, Phoebe was included in “Grupo E” by Martins and Galileo (2014a), in which the antennae are 11-segmented.

Although we have not seen a photograph of the type of Saperda phoebe / Agapanthia (Phoebe) octomaculata , the original description agrees very well with the holotype of Phoebe cretaria . All specimens, male and female, examined by us or studied through photographs have the antennae distinctly 12-segmented. Actually, it is surprising that no one has commented on this, even considering this species as the type species of Phoebe . Accordingly, this species that was wrongly considered the type species of the genus, needs to be transferred to another genus. All other species currently allocated to Phoebe have antennae 11-segmented, except Phoebe tinga Martins and Galileo, 1998 , which has antennae distinctly 12-segmented, and is transferred to Phoebella Lane, 1966 . We thus establish Phoebella tinga ( Martins and Galileo, 1998) as a new combination.

Lane (1976) described Leucophoebe for a single species, L. kempfi Lane, 1976 . According to him (translated): “It is distinguished from Phoebe Serville, 1835 (type P. phoebe ) especially by the structure and frame of the head and by the different proportions between the lengths of the prothorax and elytra. The new genus is structurally more compact.” Martins and Galileo (1998) separated Leucophoebe from Phoebe in the key (translated): “Body elongated; length of the elytra = 2.5-3 times the humeral width; elytral carina often sub-straight; tarsomeres tumid in males… Phoebe Audinet-Serville, 1835 / Body shorter; length of the elytra about twice the humeral width; elytral carina curved; tarsomeres in male not tumid…. Leucophoebe Lane, 1976 .” In the same work they transferred Amphionycha albaria Bates, 1872 (at that time Phoebe albaria ), and Phoebe pictilis Lane, 1972 to Leucophoebe . Later, Martins and Galileo (2014b) separated Phoebe from Leucophoebe in the alternative of couplet “5” (translated): “Lower eye lobes developed, longer than gena… 6 [leading to Phoebe ] / Lower eye lobes as long as gena… 7 [leading to Leucophoebe ; according to them, females of Leucophoebe cannot be separated from those of Adesmus Lepeletier and Audinet-Serville, 1825 ].” However, the body shape of the species currently included in Phoebe is variable (see photographs in Bezark 2020) and is often similar or identical to those allocated to Leucophoebe (see photographs in Bezark 2020); the elytral length is slightly longer than twice humeral width in the holotype of the type species of Leucophoebe and the holotype of L. albaria (see photograph in Bezark 2020), and almost 2.5 times in the holotype of L. pictilis (in this last case identical to that in the holotype of Phoebe fryana Lane, 1966 ) (see photograph in Bezark 2020); the shape of the protuberances on frons of males of the species of Phoebe is very variable (see photographs in Bezark 2020); the humeral carina is variable in Phoebe , and may be distinctly curved (e.g. P. cornuta ) (see photograph in Bezark 2020) or nearly straight (e.g. P. cava (Germar, 1823)) (see photograph in Bezark 2020); the lower eye lobes are very variable in the length ( Fig. 17, 19, 20, 21, 23, 25, 27 ); the shape of the metatarsomeres in males are variable in the species of Phoebe ( Fig. 18, 22, 24, 26, 28 ) and may be distinctly tumid (e.g. P. goiana Martins and Galileo, 1998 ) ( Fig. 24 ) or not tumid (e.g. P. alba ) ( Fig. 18 ). Accordingly, we were unable to find a reliable character that would allow Leucophoebe to be separated from Phoebe . Thus, Leucophoebe is considered a junior synonym of Phoebe , and therefore the following changes are proposed: Phoebe kempfi ( Lane, 1976) comb. new, Phoebe albaria (Bates, 1872) comb. new and Phoebe pictilis Lane, 1972 stat. restored.

Figures

Figures 17–28. Males. 17–18) Phoebe alba. 17) Eye and gena. 18) Metatarsus. 19) Phoebe cava, eye and gena. 20) Phoebe mafra, holotype, eye and gena. 21–22) Phoebe cornuta. 21) Eye and gena. 22) Metatarsus. 23–24) Phoebe goiana, holotype. 23) Eye and gena. 24) Metatarsus. 25–26) Phoebe bicornis. 25) Eye and gena. 26) Metatarsus.