Aedes (Ochlerotatus) caspius (Pallas)

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 : 30-34

Aedes (Ochlerotatus) caspius (Pallas)

subspecies caspius ( Pallas, 1771) —original combination: Culex caspius . Distribution: Afghanistan, Albania, Algeria, Armenia, Austria, Azerbaijan, Bahrain, Belarus, Belgium, Bosnia and Herzegovina, Bulgaria, Croatia, Cyprus, Czech Republic, Denmark, Djibouti, Egypt, Estonia, Finland, France, FYRO Macedonia, Georgia, Germany, Greece, Hungary, Iran, Iraq, Ireland, Israel ( Gaza Strip and West Bank), Italy, Kosovo, Japan, Jordan, Kazakhstan, Latvia, Lebanon, Libya, Lithuania, Malta, Mauritania, Moldova, Mongolia, Montenegro, Morocco, Netherlands, Norway, Pakistan, Palestine, People’s Republic of China, Poland, Portugal, Qatar, Romania, Russia, Saudi Arabia, Serbia, Slovakia, Slovenia, Spain (includes Balearic & Canary Islands), Sri Lanka, South Sudan, Sudan, Sweden, Switzerland, Tajikistan, Tunisia, Turkey, Turkmenistan, Ukraine, United Arab Emirates, United Kingdom, Uzbekistan, Yemen ( Robert et al. 2019; Wilkerson et al. 2021, incorrectly listed from Canada and the United States).

subspecies hargreavesi ( Edwards, 1920) —original combination: Ochlerotatus caspius var. hargreavesi (subspecific status by Harbach & Howard 2007). Distribution: Taranto, Italy [Puglia region] ( Edwards 1920).

subspecies meirai Ribeiro, da Cunha Ramos, Capela & Pires, 1980 —original combination: Aedes (Ochlerotatus) caspius meirai . Distribution: Cabo Verde [ Cape Verde] ( Republic of Cabo Verde) (Ribeiro et al. 1980).

The nominotypical subspecies was described from specimens collected in marshes near the Caspian Sea ( Pallas 1771). No type specimens are known but Edwards (1920) wrote that the species was first collected from the northern Caspian Sea in the marshes near the city of Guriev [Guryev], present day Atyrau, capital of the Atyrau Region, Kazakhstan. However, Edwards did not give a source for his information. Given that caspius is a complex of at least two species (see below), a neotype and complete description from topotypic specimens are certainly needed.

Nominal caspius can be a very abundant pest. It is distributed throughout Europe, around the Mediterranean, the Middle East, Mongolia ( Minàř 1976), China ( Becker et al. 2020) and the northern sub-Saharan countries of Chad (Ndiaye et al. 2021), Mauritania ( Mint Lekweiry et al. 2015) and Cabo Verde (subspecies meirai ). Specimens from Mauritania “showed 100% homology [rDNA ITS2] with published Ae. (Ochlerotatus) caspius sequences from Italy.” Nominal caspius is halophytic, found in dryer climates, but it is also adapted to freshwater habitats. Predictive spatial modelling by Wint et al. (2020), based on specimens and literature, reflects the above distribution, which would be expected since it is based on combined data that includes cryptic species.

The following are characters commonly used to identify the nominotypical form (after Becker et al. 2020 and Wilkerson et al. 2021).

Adult female. Hindtarsomeres 1–4 with basal and apical pale bands, hindtarsomere 5 entirely pale-scaled; wing speckled with more or less evenly mixed dark and pale scales, costa predominantly dark-scaled; abdominal coloration highly variable, terga usually with median longitudinal pale stripe, tergum VII sometimes nearly all pale, or entire abdomen pale-scaled (variability documented in Toma et al. 2017) [associated with hotter, drier areas]; scutum fawn-colored [light yellowish tan], with a pair of narrow dorsocentral pale yellowish stripes [usually referred to as white] reaching to posterior margin of the scutum.

Male genitalia. Gonocoxite with undivided basal mesal lobe (BML) [basal lobe of authors] gradually arising from the gonocoxite, bearing two closely approximated spine-like setae, larger seta hook-shaped and strongly curved apically (tip usually extending backwards to almost the middle of the seta).

Larva. Body surface without spicules; antenna uniformly spiculate, seta 1-A 3–12-branched; 18–28 comb scales in triangular patch, scales pointed (median apical spine distinctly longer than the others, at least on some scales); siphon index 1.8–2.6, elements of seta 1-S with 5–10 branches, inserted beyond middle of the siphon, with about 20 evenly spaced pecten spines reaching to about mid-length of the siphon, base of siphon with acus; segment X (anal segment) with incomplete saddle, and ventral brush (seta 4-X) with 12–17 cratal and 2 or 3 precratal setae; anal papillae lanceolate, 0.3–0.9 length of the saddle.

Edwards (1920) sought to define the nominate subspecies to separate it from closely related species, such as Ae. (Och.) dorsalis ( Meigen, 1830) and Ae. (Och.) leucomelas ( Meigen, 1804) [as salinellus Edwards, 1921b]. He described variation in a “Large series of this species [ caspius ]... recently... received at the British Museum from Italy, Macedonia, Palestine, Egypt and Mesopotamia, which show every gradation in the variation in the colour of the abdominal scales.... The thoracic coloration varies less; the two white longitudinal lines are nearly always present, though sometimes very narrow in two specimens from Italy, and in the type of G. longisquamosa [current synonym of the nominate subspecies], they are absent, the mesonotum then being uniformly fawn-coloured.” Edwards (1920) then named subspecies hargreavesi [as a variety of caspius ] stating: “Among a series sent from Taranto, Italy (E. Hargreaves), are six females which have the whole of the central area of the mesonotum covered with whitish scales though in regard to abdominal markings they agree with moderately dark specimens of the species [ caspius ]. Nothing approaching this variation has been seen from elsewhere and it therefore seems justifiable to distinguish them under a separate name.” However, it is a bit puzzling that he also mentioned the pale lines as being absent in the type of G. longisquamosa . Harbach & Howard (2007) later elevated variety hargreavesi to subspecific status in accordance with Article 45.6.4 of the International Code of Zoological Nomenclature.

The type locality of hargreavesi, Taranto , is a coastal city in southern Italy in the region forming the heel of Italy’s “boot”. We are not aware of further collections in that part of Italy and hargreavesi is usually only mentioned in lists. This leaves us without confirmation/refutation of its validity using other life stages and DNA. Townsend (1990) reported the presence of four syntypes in the collection of the Natural History Museum, London, which allows the possibility of DNA analysis.

Edwards (1921d), in a revision of Palearctic mosquitoes, wrote that the basal lobe in the closely related Ae. dorsalis was “very prominent”, as opposed to “much less prominent” in caspius . The distinction is evident in an accompanying illustration.

In Egypt, Kirkpatrick (1925) noted in a key that the basal lobe of caspius had two spines and that the appendage of the claspette was not keeled. This is in contrast to Ae. detritus ( Haliday, 1833), which has a single spine and the claspette is keeled in the middle. Among specimens that Kirkpatrick otherwise considered to be caspius , he found two larval varieties. One with “Siphon about 2.3–2.6 times as long as broad; anal gills [anal papillae] longer than broad, two or three hair tufts [setae] outside barred area [grid] of ventral brush [precratal setae]. Aedes (Ochlerotatus) caspius .” The other with “Siphon about 1.9–2.0 times as long as broad; anal gills very small and round; ventral brush with four or five tufts outside barred area. A. caspius , larval variety.”

Kirkpatrick (1925) described and illustrated the basal mesal lobe of caspius thusly: “Basal lobe rather large, triangular, with an almost straight spine [seta] rising from near its apex, and a longer stouter spine with a strongly curved tip rising at about one third of the distance from the apex; the basal lobe also bears a row of short straight bristles [setae].” A large triangular lobe and the row of short “bristles” are unlike what others had observed for caspius .

Barraud (1934) likewise described and illustrated an apparently different basal mesal lobe: “coxite [gonocoxite] with pronounced basal [mesal] lobe carrying a strong bristly spine [seta] with hooked tip, one short spine, and a small number of flattened straight, bristles [setae]”. Flattened straight setae are not mentioned elsewhere in the literature.

Marshall (1938), in The British Mosquitoes, described and illustrated all life stages (except the egg) in detail, seemingly in agreement with the general concept of caspius given above. Regarding the basal mesal lobe, he stated: “The basal [mesal] lobe of A. caspius carries a dense mass of bristles [setae], as well as two stout, conspicuous spines [setae]; the spine lying nearer to the base of the coxite [gonocoxite] being considerably longer than its companion, and having a markedly recurved extremity.As will be noted later, the hypopygium [genitalia] of A. caspius would be practically indistinguishable from that of A. dorsalis were it not for the fact that the basal lobe of the latter species is noticeably constricted at the base”. Clavero (1946) and Natvig (1948) also described similar characteristics of caspius from other European countries.

Senevet et al. (1949) reported putative caspius specimens from three sites in French North Africa: Quargla, capital city of Quargla Province in the Sahara in southern Algeria; Salah, an oasis town in central Algeria, and Gabès, capital city of the Gabès Governorate, located on the coast of the Gulf of Gabès in what is now Tunisia. The Quargla specimens were different from typical caspius and different from specimens from the other two North African sites. “The scutum is absolutely devoid of the classically described white lines. It is completely and uniformly covered with tan [pale brown] scales. The basal mesal lobe is like the classical descriptions and drawings of Edwards [1921d], of Kirkpatrick [1925], of Barraud [1934], etc., and stands out only slightly, at a slight angle, covered with numerous setae. One of them is very large and bent into an ‘S’ shape or hooked, inserted on the apical half or two thirds of the lobe. There is another less developed seta, half the length of the large one, borne anteriorly on the lobe. The lobe therefore agrees well with the illustration of Edwards” (translated from the French). Senevet et al. (1949) did not describe exactly how the basal lobe differs, but the illustration suggests a rounded lobe with a much longer primary seta. The lobes of tergum IX of males from Quargla are different from the others as well. They are twice as long as wide, straight and slightly curved inwards. The illustrations given of the ninth tergal lobes from the three sites appear different in shape and in the number and nature of the setae. The claspette is also different from others, not exhibiting a narrowing on the base of the claspette, having a large tubercle between the base and filament, and having a seta basally.

The caspius reported by Minàř (1976) from Mongolia was compared to the closely related Ae. dorsalis . Of note is the depiction of the basal mesal lobes of both species as slightly projecting but of slightly different shapes, but not rounded.

The above variation was brought into sharper focus by Cianchi et al. (1980), who, using allozyme electrophoresis, found two distinct sympatric entities, caspius species “A” and caspius species “B”. Since then there have been multiple indications of a complex of species. Darsie & Samanidou-Voyadjoglou (1997) separated two larval forms of caspius sensu lato in Greece. Schaffner (1998) suggested that only species “A” of Cianchi et al. was identified in France, so far only along the Atlantic coastline. Knio et al. (2005) wrote that “ Ochlerotatus caspius appeared to be a complex of species. Morphological differences... between the population in the South of Lebanon from the Kasmieh River and the rest of the populations. ...Kasmieh River matched the typical description of Oc. caspius ... in having... two submedial lines of whitish scales running along the length of the scutum. ...The rest of the specimens differed in having a scutum covered with scales golden-brown in color, which was similar to the description of Ochlerotatus (Ochlerotatus) mariae [scutum mainly golden brown-scaled, without definite longitudinal stripes].... Therefore, they seem to be a species near caspius and thus remain to be classified under Oc. caspius , which seems to be a species complex.” Wassim et al. (2013) stated that “They [the two forms of caspius ] have been identified as an autogenous [sic, anautogenous] form which requires a blood meal for egg development, need open space to breed (eurygamous) and abound in fresh water pools in agricultural areas, while the autogenous form is able to mate in confined spaces (stenogamous) and inhabits brackish and salty water breeding sites in coastal and inland desert areas...”. Soliman et al. (2014), in a study of egg morphology, found that the “Fine structure micrographic work of eggs of the Egyptian Ae. caspius provides new morphological evidence that both autogenous and unautogenous [sic] forms are certainly different and suggests that those forms are two distinct species.” Doosti et al. (2017) found strong evidence using mtDNA COI and COII gene sequences that indicated two species in caspius sensu lato. Detailed distributions of these two putative species have not yet been determined. Neither is it clear if caspius “A” or caspius “B” found in the Middle East corresponds to forms in Europe. The assumption seems to be that “A” and “B” are in Europe and that “A” is caspius sensu stricto.

In contrast, Porretta et al. (2011), using COI and COII gene sequences, studied the population genetics of 12 caspius populations from southern Europe and four from the western Mediterranean. They found no genetic structuring, which would be expected if habitats of this species had not been fragmented during the last ice age. They did not acknowledge the existence of caspius “A” or “B”, subspecies hargreavesi from Italy or subspecies meirai from Cabo Verde.

The existence of a species complex is not only suggested but proven with the characterization of caspius “A” and “B”, and other species probably exist in the complex. It is problematic to guess the status of subspecies hargreavesi since it has not been reported again from near its type locality and no other life stages have been studied. However, with the caveat that normal variation could be responsible for the pale scuta that Edwards observed, the character, supposedly representative of caspius “B”, by Knio et al. (2005) (see above). We tentatively predict that subspecies hargreavesi corresponds to caspius “B” of Cianchi et al. (1980). For this reason we formally elevate it here to species status: Aedes (Ochlerotatus) hargreavesi ( Edwards, 1920) . Aedes hargreavesi is not currently included as a species in the Encyclopedia of Life and should be added to the list of species of the genus.

Ribeiro et al. (1980) described subspecies meirai from islands in the Republic of Cabo Verde (also called Cape Verde). Cape Verde is located 570 km from the westernmost point of Africa, the Cape Verde Peninsula, Dakar, Senegal, after which it is named. It is in the Macaronesia ecoregion, along with the Azores, the Canary Islands, Madeira and the Savage Isles. Cabo Verde is about the same distance west from the Cape Verde Peninsula as it is from where caspius was found on continental Africa in Mauritania ( Mint Lekweiry et al. 2015).

Aedes caspius sensu lato was first recorded from Cabo Verde by de Meira (1952). Ribeiro et al. (1980) wrote: “...a detailed analysis of the Cabo-Verdian Aedes caspius revealed the existence of some morphological differences which, combined with other factors, namely the practically complete geographical isolation of the Cabo Verde population, assumed taxonomic significance at the subspecific level. ...the comparisons [with the new subspecies] were based on samples of caspius s.1. from the south of Portugal (Algarve and Alentejo)” (translated from the Portuguese). Further they wrote: “Diagnosis. Not all individuals of the new subspecies are distinguishable from the nominate subspecies…. [new paragraph] The distinction of the two subspecies of Ae. caspius is based on the presence or absence of an accessory spine [seta] on the basal [mesal] lobe of the coxite [gonocoxite], on the terminalia [genitalia] of males, duplicating the postero-external spine [seta] (fig. 59) [figure number from the original publication]. Absent in the nominal subspecies (49 specimens examined from Portugal), this spine was present, at least on one side, in 12 of the 36 males from the island of Boa Vista examined for this character (six of which, including the holotype and 2 ♂ paratypes, have accessory spines on both sides)” (translated from the Portuguese).

Ribeiro et al. (1980) also did counts of select features in comparison with their specimens from Portugal. Pupa: Setae 1-III,IV and 10-CT. Larva: Comb scales, pecten spines, antenna length to width ratio and seta 1-A. We noted significant overlap in all of these.

We do not believe that a single apparently anomalous character, present in only one third of the males of caspius from Cabo Verde, in comparison with a small sample from a single locality in the extensive range of the nominal form, is sufficient evidence to demonstrate genetic divergence. For this reason, we regard subspecies meirai as a synonym of the nominal form: meirai Ribeiro, da Cunha Ramos, Capela & Pires, 1980 , junior subjective synonym of Aedes (Ochlerotatus) caspius ( Pallas, 1771) . Consequently, “ Aedes meirai ” should be removed from the list of species of Aedes in the Encyclopedia of Life.

There is a parallel example which reinforces our decision. There are six other mosquito species in Cape Verde that are also present on mainland Africa, one of which is Anopheles (Cellia) arabiensis Patton, 1905 . A year before the description of subspecies meirai , the same authors ( Ribeiro et al. 1979) proposed a new subspecies, An. quadriannulatus davidsoni Ribeiro, da Cunha Ramos, Pires & Capela, 1979 , based on many overlapping characters with other members of the Gambiae Complex known at the time. Cambournac et al. (1982) disagreed and asserted that the species in Cabo Verde was An. arabiensis , and the populations in Cape Verde were not different from those on the mainland. The implied synonymy of davidsoni with arabiensis is currently accepted.

Eleven synonyms of the nominotypical form are listed in Wilkerson et al. (2021). Until species in the Caspius Complex are characterized, we retain, in addition to meirai , 11 synonyms of Ae. caspius : Culex punctatus Meigen, 1804 (Aden) ; Culex siculus Robineau-Desvoidy, 1827 (Sicily) ; Culex maculiventris Macquart, 1846 ( Algeria) ; Culex penicillaris Rondani, 1872 ( Italy, potentially a synonym of Ae. hargreavesi ); Grabhamia longisquamosa Theobald, 1905b ( Tunisia) ; Grabhamia subtilis Sergent & Sergent, 1905 ( Algeria) ; Mansonia arabica Giles, 1906 ( Bahrain) ; Taeniorhynchus africanus Neveu-Lemaire, 1906 ( Egypt) ; Grabhamia willcocksii Theobald, 1907 ( Egypt) ; Culex arabicus Becker, 1910 ( Yemen) ; Aedes (Ochlerotatus) epsilonn Séguy, 1924 ( Algeria) . There are no synonyms for Ae. hargreavesi ( Edwards, 1920) .