Anopheles (Cellia) rhodesiensis Theobald

Harbach, Ralph E. & Wilkerson, Richard C., 2023, The insupportable validity of mosquito subspecies (Diptera: Culicidae) and their exclusion from culicid classification, Zootaxa 5303 (1), pp. 1-184 : 70-71

Anopheles (Cellia) rhodesiensis Theobald

subspecies rhodesiensis Theobald, 1901a —original combination: Anopheles rhodesiensis . Distribution: Angola, Benin, Botswana, Burkina Faso, Cameroon, Central African Republic, Chad, Côte d’Ivoire, Democratic Republic of the Congo, Djibouti, Eritrea, Eswatini, Ethiopia, Gabon, Ghana, Guinea, Kenya, Malawi, Mali, Mauritania, Morocco, Mozambique, Namibia, Niger, Nigeria, Republic of the Congo, Republic of South Africa, Saudi Arabia, Senegal, Seychelles, Sierra Leone, Somalia, South Sudan, Sudan, Tanzania, Uganda, Zambia, Zimbabwe ( Wilkerson et al. 2021).

subspecies rupicolus Lewis, 1937 —original combination: Anopheles (Myzomyia) rupicolus (subspecific status by Mattingly & Knight 1956). Distribution: Algeria, Chad, Egypt, Eritrea, Ethiopia, Jordan, Lebanon, Morocco, Niger, Saudi Arabia, Sudan, Syria, Yemen ( Wilkerson et al. 2021).

The nominotypical subspecies was described from Mashonaland, British Central Africa, now in northern Zimbabwe. Theobald (1901a) noted that the species had been sent “in numbers” from “Central Africa”. It has since been documented throughout the Afrotropical Region ( Kyalo et al. 2017; Irish et al. 2020, who treated all records as rhodesiensis ). Theobald provided a color illustration of the habitus of the adult female ( Theobald 1901a [plates]: fig. 14, pl. IV), which matches the verbal description. Of note were: Head scales erect, dark laterally, pale on occiput and frontal tuft; scutum reddish brown with a dorsocentral ash-gray stripe; maxillary palpus with three narrow pale bands, one near apex of palpomere 2, one between palpomeres 3 and 4 and one encompassing most of palpomere 5; wing with distinct pale spots on the costa and vein R 1 as follows, sector pale (on R 1 only, illustrated by others on both veins), subcostal pale, preapical pale and apical pale spots (as defined by Wilkerson & Peyton 1990); legs entirely dark brown; abdomen brown with basal and median light brownish yellow mottling. These characters have been used in most keys to identify rhodesiensis , e.g. de Meillon (1947), Gillies & de Meillon (1968) and Gillies & Coetzee (1987). Characters of the adult, larva, pupa and male genitalia are well documented and illustrated in the three publications. According to Gillies & de Meillon (1968), larvae are found in a variety of habitats, including “rock pools, stream beds, seepages with little shade, margins of streams, springs and pools, ditches and hoof-prints, artificial containers such as concrete tanks and swimming-baths, whether in exposed water or in shade.” Females are not known to feed on humans or to be a factor in malaria transmission ( Gillies & de Meillon 1968).

Subspecies rupicolus was described from Gebel Moya near Sennar, Blue Nile Province, southeastern Sudan. Lewis (1937) remarked that the species had almost all dark wings resembling some species of the subgenus Anopheles . The color of the head scales was described as like those of subspecies rhodesiensis , with dark scales on the back of the occiput, white scales on the vertex and a conspicuous white vertical frontal tuft. The maxillary palpus and legs were described as all dark and the wing as having variably faint to absent sector pale, subcostal pale and preapical pale spots on the costa and/or vein R 1. Specimens were collected from a metal tank but most from a deep cleft in a rock. Lewis stated: “The wing markings and characters of the pharyngeal armature [cibarial armature] and male hypopygium [genitalia] indicate that the species is allied to A. rhodesiensis Theo. It differs from the latter in having 37 teeth on the mandible which is expanded at the tip, dark palpi with a higher index [length in relation to proboscis length], very faint pale marks, less than four, on the costa, no pale area on the basal third of the first longitudinal vein [R 1], and in the larva the clypeal hairs [setae 2,3-C] usually simple [single]. The mandible of A. rhodesiensis is not expanded at the end and has 80 teeth (Christophers [& Puri] 1931).” No apical pale spot was noted. De Meillon (1947) provided full descriptions of both subspecies (as species), and included illustrations showing contrasting wing spot characters and allopatric distributions ( rupicolus Saharan , rhodesiensis Afrotropical ), but then noted after the description of rupicolus : “It is questionable if this species is distinct from rhodesiensis . The adult is paler, the wings and palps [maxillary palpi] with less distinct pale making but otherwise there is little to separate them. The larvae and pupae, when considering the range of variation seen in rhodesiensis , appear to be inseparable from that species.” Mattingly & Knight (1956) cited de Meillon’s (1947) note and agreed that they too doubted the two were distinct species and provisionally preferred to treat them as subspecies. In addition, they mentioned that Lewis agreed with them. The logic and utility of this decision is not clear. However, the two nominal taxa have been treated as subspecies since that time ( Gillies & de Meillon 1968; Gillies & Coetzee 1987; Knight & Stone 1977; Harbach 2018; Wilkerson et al. 2021). Gillies & de Meillon (1968) included the two nominal taxa in their key to adult females but they are combined in the larval key as rhodesiensis . They stated that rupicolus “is not separable from the nominate form.”

An important indicator of separate species status is evidence of sympatry. Kyalo et al. (2017) did a comprehensive review of literature records of all Anopheles in the Afrotropical Region. Their dataset ( Snow 2017) was used by David Pecor (Walter Reed Biosystematics Unit) to map rhodesiensis and rupicolus . This allowed visualization of numerous instances of co-occurrence and overlapping ranges of the two nominal taxa. Also, Verrone (1962), in a paper not indexed by Kyalo et al. (2017), recorded both subspecies in close proximity in Ethiopia. The accuracy of these records is enhanced by an illustrated key to adult females clearly showing the diagnostic characters discussed above.

The above nominal taxa have been closely associated since the description of rupicolus . All stages are very similar, except for diagnostic differences in the adult female (wing with an apical pale spot, banded proboscis and many fewer mandibular teeth in rhodesiensis ). The assumption was made that differences in adult characters were due to clinal variation related to hotter temperatures in the Saharan taxon rupicolus . However, we can find no published evidence of clinal variation, and it is apparent the two forms are sympatric. This obviates the assumption of subspecific status. We therefore believe they are two genetically distinct species and formally reinstate rupicolus to its original specific status: Anopheles (Cellia) rupicolus Lewis, 1937 . Anopheles rupicolus is currently listed as a species in the Encyclopedia of Life.

Anopheles rupicolus currently has two synonyms, which we retain: An. aegypti Salem, 1938 (type locality: Wadi Taba, Sinai, Egypt) and An. rhodesiensis var. dthalisimilis Corradetti, 1939 (type locality: Semien District, Ethiopia). Both were treated as questionable synonyms by Edwards (1941), which leaves some doubt about their status and lays open the possible existence of a species complex.