Given insects' incredible variability in morphology, physiology, and behavior, they have long been the focus of studies examining the processes that generate and maintain biological diversity (reviewed in Price 2003). More recently, numerous insect groups have been used to examine the patterns of genetic variation that underlie phenotypic differences, as well as to explore processes of speciation and radiation. Examples of such groups include the Drosophila of Hawaii (reviewed in O'Grady and DeSalle 2018), the soapberry bugs of Florida (e.g., Carroll and Boyd 1992, Carroll et al. 2003, 2012), and the tiger swallowtail butterflies of Appalachia (Ording et al. 2010, Kunte et al. 2011). These and numerous other insect taxa have allowed entomologists to test hypotheses of evolutionary change, taxon formation, and the maintenance of reproductive isolation within and between species, processes that broadly affect biodiversity. This research also has practical applications, specifically in the development of insect conservation or control strategies (e.g., Aardema et al. 2011, Gourbière et al. 2012, Buckley 2016, Schmidt et al. 2021).

Although numerous insect systems have been used to examine taxonomic radiation and evolution, in most instances, the processes observed have occurred (or are occurring) on a relatively local or regional geographic scale, including the examples listed above. In contrast, patterns of evolution in the Culex pipiens L. (Diptera: Culicidae) complex of mosquitoes likely occurred broadly across the globe as a result of their associations with humans. This suggests that studies of evolutionary processes in this system may reveal unique or additional patterns of diversity formation and maintenance that are absent in more geographically restricted taxa. Furthermore, some of the species in this complex are worldwide invaders that can transmit several pathogens of medical significance (Turell 2012), which gives additional importance to future work on these mosquitoes.

The Cx. pipiens complex has historically represented 'one of the major outstanding problems in mosquito taxonomy' (Harbach et al. 1985). This is because a remarkable array of divergent and epidemiologically significant physiological and behavioral traits occurs on a background of extremely low morphological diversity, stymieing generations of medical entomologists trying to name and understand these mosquitoes. However, advances in genetic and genomic sequencing (i.e., next-generation sequencing technologies) have facilitated recent phylogenetic studies within this group (e.g., Aardema et al. 2020, Yurchenko et al. 2020, Aardema et al. 2021), and the evolutionary history of these mosquitoes is now coming into focus.

We propose that the foundational knowledge and necessary genomic tools are presently available to make the mosquitoes that form the Cx. pipiens complex an excellent system for examining broad evolutionary processes. In particular, we argue that this complex of closely related taxa is an exceptional system to study the mechanisms underlying speciation and taxonomic radiation, as well as the influences that historical and contemporary admixture have on taxonomic integrity. Here, we review the ecological and behavioral diversity seen within the complex and discuss the evolutionary patterns uncovered by recent genetic and genomic studies. We also discuss future avenues of inquiry that will broaden our understanding of these Culex mosquitoes specifically, as well as organic evolution, adaptation, and speciation in insects more generally. Finally, we consider the important ramifications future studies may have for management of these mosquitoes, work that could lessen the substantial disease and nuisance burdens they cause.

Currently, there are four recognized species in the Cx. pipiens species complex (also referred to as the pipiens subgroup [Wilkerson et al. 2021], see Table 1). These are the globally distributed Cx. pipiens and Cx. quinquefasciatus, as well as Cx. australicus and Cx. globocoxitus, two species endemic to the Australian continent and Tasmania. The complex also includes a recognized subspecies, Cx. pipiens pallens. Finally, Cx. pipiens has two ecologically distinct forms: Cx. pipiens f. pipiens and Cx. pipiens f. molestus. These two forms have no consistent morphological differences as adults (Harbach et al. 1984), but vary considerably in several important ecological and behavioral aspects (discussed below). The nature of the 'molestus' form has been a specific topic of intense consideration and debate (reviewed in Harbach 2012). Most relevantly, it was historically unclear whether Cx. pipiens f. molestus represented a single evolutionary entity that evolved once and then expanded its range through human-mediated processes (Fonseca et al. 2004, Gomes et al. 2015, Yurchenko et al. 2020), or conversely whether it is a local variant of Cx. pipiens that evolved repeatedly, and independently under specific conditions such as those found in areas of dense human habitation (e.g., Byrne andNichols 1999, Kothera et al. 2010).

The taxa within the Cx. pipiens complex were historically differentiated morphologically, and as there are many good discussions of these reported differences (e.g., Vinogradova 2003, Harbach 2012), we will not endeavor to further review them here. More recent genetic analyses, using a variety of molecular markers (e.g., protein variation, mitochondrial DNA regions, microsatellites, and autosomal genetic markers), have broadly revealed taxonomic designations and distinctions within the complex (e.g., Miles and Paterson 1979, Fonseca et al. 2004, Weitzel et al. 2009). However, these studies have also revealed the limitations of small amounts of molecular data, with many analyses being unable to resolve taxonomic relationships among complex members (e.g., Miller et al. 1996, Koosha et al. 2017). Mitochondrial markers, in particular (including whole mitochondrial genomes), have presented challenges for confident phylogenetic analysis (Laurito et al. 2013, Luo et al. 2016, Batovska et al. 2016, da Silva et al. 2020). One possible reason for this difficulty is that infection with the endosymbiont Wolbachia has strongly affected patterns of maternal inheritance of mitochondrial markers throughout the Cx. pipiens complex in ways that do not align with taxonomic designations (Dumas et al. 2013).

The four recognized species are listed, as well as the subspecies and ecologically forms of Cx. pipiens. All taxonomic designations follow those given in Wilkerson et al. 2021. Also included is information pertaining to an inferred last common ancestor of the clade. For several of the taxa, there is some ambiguity regarding the associated blood hosts due to conflicting reports as well as lack of investigation. Seasonality, specifically whether a taxon is heterodynamic or homodynamic, also likely varies to some extend in several of the complex taxa. For example, although Cx. globocoxitus is typically reported to be homodynamic (e.g., Russell 2012), its presence in Victoria and Tasmania suggests that it likely has adaptations for seasonal climates in at least some locations. Accordingly, it may be heterodynamic in these locations. See text for more details.

More recent studies, making use of large numbers of nuclear markers, have been applied to this complex, and the evolutionary relationships between complex members have become better resolved (e.g., Aardema et al. 2020, Yurchenko et al. 2020, Aardema et al. 2021). In particular, general phylogenetic relationships and patterns of divergence have become clearer, although much work still remains in this area. Considering these phylogenetic relationships, recent genetic studies have suggested that the two Australian endemic species are most closely aligned with one another (Batovska et al. 2016, Aardema et al. 2020), supporting results found from comparisons of protein variation (Miles and Paterson 1979). These two species may be sister to Cx. quinquefasciatus, and collectively these three taxa are then sister to Cx. pipiens (Fig. 1). However, it is possible that genetic exchange between Cx. quinquefasciatus and one or both of the Australian endemic taxa may be confounding these results (Aardema et al. 2020, D. M. Fonseca, unpublished data). Such reticulated evolution notoriously makes determining phylogenetic relationships difficult (Mallet et al. 2016) and may be another reason why resolution of divergence patterns within the Cx. pipiens complex has been so difficult.

Genetic exchange has also created challenges for determining the extent to which Cx. pipiens f. pipiens and/or Cx. pipiens f. molestus represent monophyletic groups. However, with recent phylogenetic studies, it has become clear that populations of Cx. pipiens f. molestus rarely, if ever, represent locally diverged forms of Cx. pipiens f. pipiens that have adapted to urban conditions (Fonseca et al. 2004, Weitzel et al. 2009, Yurchenko et al. 2020). Rather, contemporary populations of Cx. pipiens f. molestus are likely derived from a single evolutionary event.

Regarding the subspecies Cx. pipiens pallens, it is unclear where it lies in a phylogenetic context. Some studies place it more closely aligned with Cx. quinquefasciatus (e.g., Kasai et al. 2008, Aardema et al. 2020), whereas others have it as more similar genetically to Cx. pipiens (e.g., Luo et al. 2016, da Silva et al. 2020, Aardema et al. 2021). This discrepancy may not be particularly surprising as genetic and genomic analyses suggest that Cx. pipiens pallens could have arisen through ancestral hybridization between Cx. quinquefasciatus and an unknown taxon that may have been most genetically similar to contemporary Cx. pipiens (Fonseca et al. 2009, Aardema et al. 2020).

One of the major distinguishing characteristics among the taxa within the Cx. pipiens complex is the degree to which they are adapted to seasonal climates and the amount of variation in activity throughout the year (i.e., active year-round [homodynamic] or active seasonally [heterodynamic]; see Table 1). Culex pipiens f. pipiens and Cx. pipiens pallens are considered heterodynamic and are predominantly found at northern latitudes of temperate North America, Europe, and Asia. In these locations, shorter photoperiods and decreasing temperatures experienced by juveniles during late summer and autumn induce reproductive diapause in adult females (Eldridge, 1968;Spielman and Wong, 1973). Like many diapausing insects, these mosquitos store lipids prior to entering diapause (Zhou and Miesfeld, 2009) and then find sheltered locations where they remain throughout the winter (Vinogradova, 2000). Males cannot diapause and die with the onset of winter. The ability to diapause is an important adaptation that allows these mosquitoes to thrive in northern temperate climates. However, Cx. pipiens f. pipiens is also found throughout southern Europe as well as northern and sub-Saharan Africa, indicating that it can persist in warmer locations. Interestingly, a Cx. pipiens f. pipiens population from South Africa appeared incapable of entering diapause (Jupp 1987). These mosquitoes could represent an ancestral form of Cx. pipiens in Africa (discussed below). Further evidence in support of this hypothesis comes from the observation that a large portion of Cx. pipiens from South Africa lack the endosymbiont Wolbachia (Cornel et al. 2003, Rasgon et al. 2006). It is generally presumed that an absence of infection is the ancestral state (Miles and Paterson 1979).

The second, 'molestus' form of Cx. pipiens appears to be a taxon adapted to warmer climates (Harbach 1984). It is homodynamic and incapable of entering diapause. However, several adaptations have allowed this mosquito to colonize and thrive underground in urban areas at colder latitudes as far north as Siberia (Vinogradova 2000). Chief among these adaptations are an ability to readily mate in enclosed spaces ('stenogamy') and an ability to lay eggs without a bloodmeal ('autogeny'). Genetic divergence contributing to the distinct differences in diapause ability between the two Cx. pipiens forms has received some attention (e.g., Epstein et al. 2021), and ongoing research in this area may facilitate a greater understanding of seasonal adaptations in mosquitoes more generally.

The two Australian endemic species, Cx. australicus and Cx. globocoxitus, are also generally considered to be adapted to temperate climates (Dobrotworsky 1967a), although the northern range of Cx. australicus extends into subtropical regions (Webb et al. 2016). Neither of these species has been reported to undergo reproductive diapause, however, although Cx. australicus can enter a quiescent state during inclement weather (Dobrotworsky andDrummond 1953, Dobrotworsky 1967a). Correspondingly, it is considered to be heterodynamic. Minimal information has been reported regarding potential seasonality and corresponding adaptations in Cx. globocoxitus. It is stated to be homodynamic (Russell 2012), although its range extends into Tasmania (Dobrotworsky 1966), suggesting that it may harbor some as yet unknown mechanisms to persist in seasonal, temperate environments.

Culex quinquefasciatus is considered to be a tropical or subtropical species (Fonseca et al. 2006, Harbach 2012). Its historical range likely extended from eastern sub-Saharan Africa west to southeast Asia. Genetic evidence suggests that human migration facilitated recent expansion of this mosquito into Australia, Polynesia, North and South America, and western Africa (Fonseca et al. 2006). Throughout both its native and introduced ranges, Cx. quinquefasciatus is homodynamic and generally lacks an ability to respond to climatic seasonality (e.g., diapause or quiescence).

Adult females of all Cx. pipiens complex taxa will take blood from birds ('ornithophily'), and this seems to be the ancestral blood source preference of this clade (Barr 1967). However, Cx. pipiens f. molestus may obtain blood meals predominately from mammalian blood meals or even specifically humans (Harbach et al. 1984), but the extent to which these may be preferences (i.e. 'mammophily' and/or 'anthropophily'), versus simply differences in access to potential hosts remains in need of further study (Gomes et al. 2013). Culex quinquefasciatus is also known to readily take blood meals from mammals (including humans), but there is substantial variation in the strength of this behavior between populations (e.g., Samuel et al. 2004, Molaei et al. 2007, Muturi et al. 2008, Garcia-Rejon et al. 2010). Culex globocoxitus and Cx. australicus are broadly considered ornithophilic, although both have also been reported to obtain blood meals from mammals, specifically rabbits (Lee et al. 1954). The degree to which either taxon may exhibit more general mammalophilic tendencies would benefit from additional investigation.

Intraspecific mating of adult Cx. pipiens f. molestus, Cx. quinquefasciatus, Cx. globocoxitus, and Cx. pipiens pallens can occur in enclosed spaces ('stenogamy'), allowing them to be colonized. In contrast, Cx. pipiens f. pipiens and Cx. australicus mate in large swarms that congregate in open spaces ('eurygamous') and are correspondingly difficult to maintain in a lab. Eurygamy has been proposed as the ancestral mating mode for the complex (Barr 1967), as it is the state predominately observed in closely related Culex taxa. Stenogamy, as well as mammophily, may be derived characteristics, representing relatively recent adaptations in conjunction with closer affiliations with human habitation ('domesticity', Dobrotworsky 1965).

Higher elevational regions such as those found in Ethiopia may have contributed to the evolution of contemporary Cx. pipiens (Fig. 1; Ross 1953, Barr 1967), as these could have selected for heterodynamy and possibly reproductive diapause. From east Africa, these Cx. pipiens likely spread throughout much of the Old-World through natural expansion processes, further adapting to cooler climates in Europe and Asia to become the modern form of Cx. pipiens f. pipiens. Currently, the best supported hypothesis posits that Cx. pipiens f. molestus evolved in the Mediterranean or Near East, possibly congruent with more stable human settlements and increasing urbanization (Fig. 1; Fonseca et al. 2004). Interestingly, Cx. pipiens f. molestus was first described from the Middle East (specifically Egypt; Forskål 1775), and contemporarily, this appears to be the dominant form of Cx. pipiens in this region (Villani et al. 1986, Nudelman et al 1988, Mohsen et al. 1995).

In contrast to previous reports (e.g., Byrne & Nichols 1999, Kothera et al. 2010), Cx. pipiens f. molestus is unlikely to have evolved from a Cx. pipiens ancestor that resembled contemporary Cx. pipiens f. pipiens found in temperate climates. Rather, it likely derived from a more ancestral Cx. pipiens population that lacked an ability to enter reproductive diapause (Fonseca et al. 2004). Some contemporary populations of Cx. pipiens f. pipiens in sub-Saharan Africa are incapable of diapausing (Jupp 1987), and it is possible that these are remnants of a more ancestral Cx. pipiens form.

Culex quinquefasciatus may have evolved in east Africa as well, or else in southeast Asia (Fig. 1; Fonseca et al. 2006). Contemporary samples from these locations are genetically similar and have high levels of genetic diversity, hindering more refined considerations of geographic origins. Regardless, from its source location, it has subsequently radiated outward, most likely associated with human migration, both historically and ongoing (Bataille et al. 2009).

Given their contemporary ranges, Cx. australicus and Cx. globocoxitus likely evolved somewhere in temperate Australia (Fig. 1), possibly from a shared common ancestor that may have resembled an ancestral form of Cx. quinquefasciatus (Dobrotworsky 1967a). Current populations of Cx. quinquefasciatus in Australia may have been introduced relatively recently, likely after the discovery of the continent by Europeans (Russell 2012). This possibility is supported by their genetic similarity to other Asian populations (Fonseca et al. 2006). Interestingly, the Australian endemic taxa lack the endosymbiont Wolbachia (Irving-Bell 1974). This observation has been interpreted as evidence for their relative distinctiveness and early evolutionary divergence from the rest of the complex (Miles and Paterson 1979).

Finally, Cx. pipiens pallens may represent a taxon derived from hybridization between Cx. quinquefasciatus and Cx. pipiens somewhere in east Asia where it is currently distributed (Fig. 1; Fonseca et al. 2009, Aardema et al. 2020). The range of Cx. quinquefasciatus extends from tropical southeast Asia into southern China, Korea, and Japan where contemporarily it overlaps with southern populations of Cx. pipiens pallens. However, when considering what form of Cx. pipiens may have contributed to the genesis of Cx. pipiens pallens, many questions arise. Culex pipiens f. molestus is found in urban areas overlapping with the current range of Cx. pipiens pallens. However, these are considered to be recent introductions, possibly around or during World War II (Drummond 1951, Mogi 2012). Furthermore, Cx. pipiens pallens has an ability to enter reproductive diapause, a trait which both Cx. pipiens f. molestus and Cx. quinquefasciatus lack. Conversely, Cx. pipiens f. pipiens as the second parental taxon contributing to the formation of Cx. pipiens pallens also presents challenges. Specifically, there are no known populations of Cx. pipiens f. pipiens in proximity to the current range of Cx. pipiens pallens (but see Gao et al. 2016). However, considering how substantially the Pleistocene glaciations (2.58 to 0.18 million years ago [Gibbard et al. 2010]) affected insect distributions (Coope 1994, Hofreiter andStewart 2009), it is possible that they decimated the parental temperate Cx. pipiens-like northern populations, and after their retreat provided an open niche for hybridization-derived Cx. pipiens pallens populations formed in southern zones of overlap.

The last common ancestor (LCA) of the Cx. pipiens complex may have first evolved in east Africa (Fig. 1; Ross 1953, Barr 1967, Vinogradova 2000, Fonseca et al. 2006). It likely did not fully match any specific, contemporary taxon (Barr 1967). Like Cx. quinquefasciatus it was probably homodynamic. However, like Cx. pipiens f. pipiens it was likely predominantly ornithophilic and eurygamous (Barr 1967). Adaptations to seasonal climates (e.g., reproductive diapause or quiescence), expansion of recognized blood hosts to include mammals (including humans), and the ability to mate in enclosed spaces are all likely to be derived traits that evolved in contemporary members of the complex as they adapted to temperate climates and/or greater affiliation with human settlements.

The underlying proximate mechanisms contributing to evolutionary divergence within the Cx. pipiens complex is a subject that has received scant attention, and the factors facilitating the emergence of the four recognized Cx. pipiens species remain unclear. For example, none of the four appears to exist contemporarily in geographic isolation (Harbach 2012). However, much of this is clearly due to human-mediated movement of mosquitoes (Fonseca et al. 2006), such as in the Western Hemisphere where Cx. quinquefasciatus and Cx. pipiens overlap across much of the United States, or in Australia and east Asia where Cx. pipiens f. molestus is believed to have been introduced during World War II (Drummond 1951, Mogi 2012). Culex pipiens is hypothesized to have evolved in east Africa, possibly in higher elevational regions (Barr 1967). Such regions may have selected for adaptations, such as heterodynamy, that allowed this species to ultimately colonize temperate locations across the globe. East Africa has also been proposed as the geographic origin of Cx. quinquefasciatus, although it is also possible that it arose in tropical Asia (Fonseca et al. 2006). In the latter case, this would suggest that modern forms of Cx. pipiens and Cx. quinquefasciatus could have evolved in allopatry. If modern Cx. quinquefasciatus first evolved in east Africa, it is also possible that differences in elevational preferences (i.e., Cx. quinquefasciatus at lower elevations and Cx. pipiens at higher elevations) may have allowed each taxon to evolve along independent lines.

In Australia, the contemporary distributions of Cx. quinquefasciatus, Cx. australicus, and Cx. globocoxitus overlap in southern (temperate) regions, although Cx. quinquefasciatus is likely absent from the coldest latitudes (e.g., Victoria and Tasmania). Likewise, Cx. australicus and Cx. globocoxitus are absent in more northern, tropical locations (Webb et al. 2016). Culex australicus and Cx. globocoxitus overlap extensively with one another, although Cx. australicus is reported to be more widespread, extending north into subtropical regions of Australia (Dobrotworsky and Drummond 1953). It is possible that differences in mating behaviors (stenogamy in Cx. globocoxitus versus eurygamy in Cx. australicus) may contribute to reproductive isolation between them. As research into these two species has been limited to date, more work will be needed to assess spatial and/or temporal factors that may have led to their evolution as distinct species.

Finer scale separation created by divergence in preferred habits may also be a contributing factor in the maintenance of some isolation between Cx. pipiens f. pipiens and Cx. pipiens f. molestus. The former is found almost always aboveground, whereas in more temperature locations, Cx. pipiens f. molestus develops exclusively in urban, underground locations such as sewers, flooded basements, and subway tunnels (reviewed in Vinogradova 2000). Variation in mating behaviors may also facilitate some pre-zygotic reproductive isolation between the two Cx. pipiens forms. Nonetheless, evidence of genetic exchange between the two forms has been found in numerous studies (e.g., Chevillon et al. 1998, Fonseca et al. 2004, Gomes et al. 2009) and is strong evidence that these mosquitoes do frequently interbreed where their distributions overlap.

Regardless of whether historical vicariance facilitated taxonomic radiation within the Cx. pipiens complex, it is clear that adaptation to varying ecological conditions has been a driver of divergence among these mosquitoes. In particular, the ability to diapause in response to inclement seasonal weather may be a major factor contributing to the maintenance of discreet taxa. This is because an inability to diapause may limit the distribution of Cx. quinquefasciatus to warmer climates, especially in its historical range where Cx. pipiens f. pipiens is absent. Where Cx. pipiens f. pipiens and Cx. quinquefasciatus do overlap (e.g., North America), hybridization between them could theoretically facilitate adaptive introgression of the diapause trait into Cx. quinquefasciatus populations. This, in turn, could allow Cx. quinquefasciatus to expand its range northwards. Future monitoring will be required to assess whether this is occurring.

It is also likely that an inability to diapause limits the distribution of Cx. pipiens f. molestus to urban areas in temperate locations. Conversely, the diapause trait could also constrain expansion into warmer climates by Cx. pipiens f. pipiens and Cx. pipiens pallens. At lower latitudes where there is less seasonal variation in daylength, there can be temporal mismatches between the thermal environment experienced by individuals and the cues (such as photoperiod) that have evolved to induce diapause (Kellermann and van Heerwaarden 2019). Although this is unlikely to prevent population expansion over evolutionary timescales, such mismatches could limit shorter term expansion potential due to increases in environmental stress, predation, disease susceptibility, and/or competition.

Hybridization Leading to New Taxa?

Perhaps, one of the most intriguing phylogenetic aspects of the Cx. pipiens complex is the possibility that genetic exchange (i.e., interspecific hybridization) has led to unique taxonomic entities (Fig. 1). For example, recent analyses support the hypothesis that the subspecies Cx. pipiens pallens shares substantial ancestry with both Cx. pipiens and Cx. quinquefasciatus (Fonseca et al. 2009, Aardema et al. 2020, Aardema et al. 2021). Crossing studies among members of the Cx. pipiens complex showed that the nuclear gene acetylcholinesterase-2 (ace-2) is physically linked to the autosomal male determining locus (MDL), which is found only in male mosquitoes (Malcolm et al. 1998). The sequence of the ace-2 gene is divergent between Cx. quinquefasciatus and Cx. pipiens pallens (Smith and Fonseca 2004), and for females of both taxa, this region is both homozygous and taxonomically diagnostic. However, males of Cx. pipiens pallens in populations with no evidence of ongoing hybridization with Cx. quinquefasciatus have two distinct ace-2 sequences (i.e., heterozygosity). One of these sequences matches the 'pallens' form of ace-2, whereas the other is similar to that of 'quinquefasciatus'. Females from these same populations only have the 'pallens' sequence (Fonseca et al. 2009). There is no genomic recombination around the MDL, even though the rest of the genome freely recombines. These observations suggest that Cx. pipiens pallens originated from unidirectional hybridization between male Cx. quinquefasciatus and females from a diapausing form of Cx. pipiens (Fonseca et al. 2009). The lack of recombination means that this DNA region has remained unchanged and correspondingly retains this signature of its Cx. quinquefasciatus ancestry.

Contemporarily, Cx. pipiens pallens harbors substantial amounts of unique genetic variation, suggesting that it is not simply a hybrid swarm, but rather its own, unique evolutionary entity (Fonseca et al. 2009, Aardema et al. 2020, Aardema et al. 2021). The hypothesis that Cx. pipiens pallens arose by hybridization still requires substantial further inquiry, but the recent publication of a high quality Cx. pipiens pallens reference genome will no doubt aid these efforts (Peng et al. 2021). Investigation to formally examine the extent and consistency of admixture across the genome, and across the range of Cx. pipiens pallens, will be very informative.

Further considering the ace-2 locus, a similar pattern to that described above was subsequently observed in Cx. australicus (D. M. Fonseca, unpublished data). Specifically, females have a unique ace-2 sequence (Smith & Fonseca 2004), whereas males harbor both this sequence and one identical to that of Cx. quinquefasciatus. Interestingly, mitochondrial DNA sequences of Cx. globocoxitus are most similar to those of Cx. australicus (Batovska et al. 2016, da Silva 2020). These results suggest that male Cx. quinquefasciatus may have mated with female Cx. globocoxitus, resulting in a hybrid taxon that is now identified as Cx. australicus. The appearance of the male genitalia of Cx. australicus (like that of Cx. pipiens pallens) resembles the genitalia of lab-produced hybrids between Cx. pipiens and Cx. quinquefasciatus (Vinogradova 2000). This observation also supports the hypothesis of a hybrid origin and further underscores the need for future analyses.

Many early investigation of Cx. pipiens complex members focused on the extent and nature of pre-zygotic reproductive isolation. In particular, examinations of assortative mating were common and showed that under many contexts, populations and taxa appear to preferentially mate with conspecifics (Dobrotworsky 1967b, Mattingly 1967). However, with the advent of genetic analyses, it has become increasingly clear that introgression between taxonomic members is far from infrequent (e.g., Chevillon et al. 1998, Fonseca et al. 2009, Gomes et al. 2012, Aardema et al. 2020). Indeed, one of the most common and challenging observations made within the complex is the extent of genetic exchange between seemingly distinct populations and taxa.

One of the clearest and possibly consistent regions of genetic introgression in the complex occurs between Cx. pipiens and Cx. quinquefasciatus in the United States. As a species adapted to temperate regions, Cx. pipiens f. pipiens is found at higher latitudes in North America, whereas the more tropical-adapted Cx. quinquefasciatus is limited to lower latitudes. Samples taken from the central states in the United States are almost always hybrids between the two species (Joyce et al. 2018). This hybrid zone could be used to examine hypotheses of species maintenance in the face of genetic exchange (e.g., Kothera et al. 2009) and could also aid in trait mapping and examinations of differential selection in these species (Lexer et al. 2004, Turner and Harr 2014, Brelsford et al. 2017). Lab-based hybrids between Cx. quinquefasciatus and Cx. pipiens have already been used to successfully develop quantitative trait loci (QTL) for the diapause trait (Mori et al. 2007).

Another important aspect of hybridization in this complex is between the two forms of Cx. pipiens. This importance derives from the hypothesis that hybrids of the two Cx. pipiens forms may exhibit blood feeding preferences that includes both birds and humans, facilitating the transmission of West Nile virus from its natural bird reservoirs to human populations (Fonseca et al. 2004, Farajollahi et al. 2011). Nearly every study that has investigated genetic exchange between the two forms has found evidence of such introgression (e.g., Chevillon et al. 1998, Fonseca et al. 2004, Kent et al. 2007, Gomes et al. 2009, Amraoui et al. 2012, Beji et al. 2017), and it is clear that these mosquitoes frequently encounter one another and reproduce where their distributions overlap, predominately in urban areas. Hybrids of these two forms may also facilitate investigations of divergent blood host preferences in this complex (Fritz et al. 2015).

Finally, there appears to be substantial contemporary introgression from Cx. quinquefasciatus into Cx. pipiens pallens, especially in more southern parts of Cx. pipiens pallens' geographic range (Fonseca et al. 2009, Gao et al. 2016, Aardema et al. 2020). Natural hybridization between Cx. pipiens pallens and Cx. pipiens f. molestus has additionally been reported from Japan (Ohashi et al. 2014), although this same study also noted fitness reductions in hybrids. As Cx. pipiens f. molestus is likely to be a recent introduction in East Asia, it is possible that pre-zygotic isolating mechanisms between these taxa could be actively evolving as a result of these fitness reductions in hybrids.

In addition to examinations of inter-taxonomic patterns of genetic diversity and introgression, studies have also revealed high amounts of genetic diversity harbored within populations and taxa (e.g., Smith et al. 2005, Fonseca et al. 2006). These high levels of genetic diversity inherently complicate investigations of taxonomic divergence between members of the complex. Conversely, this large amount of genetic variation can be explicitly utilized to further understand patterns of gene flow and the global expansion of this complex (e.g., Fonseca et al. 2006, Asgharian et al. 2015, Aardema et al. 2021). Such analyses will be necessary to better understand historical distributions and the extent of genetic exchange between these Culex populations.

Recent developments in nuclear genomic data acquisition (e.g., next-generation sequencing), combined with powerful analytical tools and phylogenetic methodologies, have created a renaissance in studies of insect phylogenomics and evolution (e.g., Misof 2014, Kawahara and Breinholt 2014, Johnson et al. 2018, Kawahara et al. 2019). Large, informative datasets can do more than simply elucidate relative relationships between taxa; they can also be used to form a better understanding of how genes and genomes evolved in response to specific demographic events, ecological changes, or other selection pressures (e.g., Zhen et al. 2012, Logue et al. 2015, Brand et al. 2018, Veale et al. 2018). In short, we are now able to identify correlations between specific selective forces or demographic events and underlying genomic diversity. This can subsequently allow entomologists to better understand insect biodiversity in rich and detailed ways that were previously unobtainable.

As we have outlined above, the Cx. pipiens complex has a wealth of biological properties that make it an excellent system for exploring questions of taxonomic divergence and adaptive evolution. In combination with these properties, considerable genetic and genomic resources have been developed for members of the complex. The first assembled genome from a complex member, Cx. quinquefasciatus, was published in 2010 (Arensburger et al. 2010), and more recently, it was improved with the use of Hi-C data to a chromosomal level (Dudchenko et al. 2017). A genome from Cx. pipiens pallens was also recently made available (Peng et al. 2021). Furthermore, the availability of an assembled genome for the closely related mosquito, Cx. tarsalis (Main et al. 2021), provides an outgroup species for the complex that could facilitate a better understanding of the evolutionary origins of traits and taxa.

As well as these assembled genomes, several studies have also produced whole genome sequencing data for single samples of other complex members (Aardema et al. 2020, Yurchenko et al. 2020) or from pooled samples (reviewed in Aardema et al. 2020). Further studies have examined patterns of exonic differentiation between complex taxa (e.g., Price and Fonseca 2015) or gene expression patterns under differing conditions (e.g., Leal et al. 2013, Honnen et al. 2016, Li et al. 2016, Hagan et al. 2018). These datasets are available and could be utilized for further analyses of trait evolution and genetic comparisons within the complex. However, these data alone will be insufficient to answer the truly large and important questions found within the Cx. pipiens complex and one of the greatest challenges moving forward will be in identifying and obtaining appropriate samples. It will also become necessary to move beyond analyses of correlative relationships between genetic variation and phenotypes of interest to experimental validation of these observations (Ruzzante et al. 2019). RNA interference (RNAi) work (Wu et al. 2020), and the recent application of CRISPR technology to Cx. quinquefasciatus (Feng et al. 2021), shows that such research is contemporarily possible.

One of the most pressing needs for a better understanding of the evolutionary history and taxonomic relationships within the Cx. pipiens complex is an increased focus on regions underrepresented in previous analyses. Africa, in particular, is critical for understanding early evolutionary events in the radiation of the complex. Culex pipiens populations in sub-Saharan Africa should be examined to further our understanding of how this species ultimately adapted to temperate climates. As this is a region where Culex diversity has to date been woefully understudied, the exact makeup of these mosquitoes remains to be determined, both ecologically and genetically. Despite harboring a number of complex taxa, Australia has also received surprisingly little recent attention. More sampling and genomic analyses from this continent would provide significant opportunities to further our knowledge of evolution in Culex mosquitoes.

One of the most pressing reasons for understanding taxonomic relationships, common evolutionary processes, and patterns of genetic exchange within the Cx. pipiens complex is to develop strategies to reduce or prevent its members from transmitting infectious diseases to humans and/or domestic animal populations. Research suggests that among and within the species of the complex, there is a variation in their vector capabilities (Turell et al. 2014). How these differences are manifested through genetic variation (relative to environmentally influenced behavioral or ecological differences, for example) remains to be explored. The degree to which taxonomic background influences host feeding preferences and therefore potential vector capabilities has received some attention, particularly with consideration to the two forms of Cx. pipiens (e.g., Kilpatrick et al. 2007, Huang et al. 2009, Osório et al. 2012, Fritz et al. 2015). However, substantial work remains to fully understand how these patterns are manifested at global scales.

Another reason why this research is critical is that genetic exchange promotes the spread of allelic variation that confers resistance to insecticides (Guillemaud et al. 1996, Chevillon et al. 1998). Formal and thorough examinations of gene flow among populations and between taxa will provide the foundation needed to make predictions about when and where resistance may arise, as well as help identify resistance alleles already present in contemporary populations (e.g., Aardema et al. 2021).

Finally, an important area of future research is the possibility that ancestral hybridization between distinct Culex species has led to novel and distinct taxa. That genetic exchange between divergent taxa can form novel evolutionary entities has become increasingly appreciated (e.g., Kunte et al. 2011, Lamichhaney et al. 2018), but examples of this process remain few. Accordingly, efforts to understand the relationship between hybridization and taxon formation have been limited. The identification of additional taxa that may harbor hybrid origins in their evolutionary history may provide new insights into so called 'hybrid speciation'. As discussed above, Cx. pipiens pallens and Cx. australicus may be such examples, although such origins remain hypotheses requiring additional research. Regardless, it is clear that the Cx. pipiens complex may offer opportunities to explore hybrid taxon formation in a system with a wealth of available ecological and genetic resources and with significant public health implications.

The Cx. pipiens complex has a long and rich history of scientific attention, but has continuously presented challenges for taxonomists and those wishing to understand the evolutionary history of these mosquitoes. Recent advances in molecular tools and techniques are now being applied to answer such questions, revealing complicated genetic patterns and taxonomic relationships. These investigations are also illuminating the potential of these mosquitoes to address fundamental questions in evolutionary biology. Such investigations are critical for making predictions about future evolutionary trajectories in response to changing environments in this complex and may inform studies of evolutionary processes in other invasive species of medical or economic significance.