Populations of Culex stigmatosoma and Cx. thriambus have been documented in the southwestern USA with a southward range extension to northern South America and Central America, respectively. Studies conducted in California indicate both species are potential vectors of West Nile virus. However, vector competence studies are lacking for other parts of the USA. During a multicounty regional surveillance study west of San Antonio, Texas, multiple errors were observed in the Texas distributional literature of these species. These errors involved incorrect distributional information in Texas and US publications. Evidence to correct these errant records was found upon further analysis of Texas literature and curated specimens. Therefore, the aims of this study were to present that evidence and then combine the corrected records with additional records from the Texas Department of State Health Services and from larval collections made during other Texas surveillance studies.
Knowledge about the spatial distribution of vector species can be used to predict disease risk and inform vector control activities. Culex stigmatosoma Dyar and Culex thriambus Dyar have been indicated as potential West Nile virus vector species based upon vector competence experiments in California (Goddard et al. 2002, Reisen et al. 2006). Information is lacking on their geographic distributions and how these species may affect disease transmission throughout the southwestern USA. West Nile virus is the most prevalent mosquito-borne pathogen in the USA (McDonald et al. 2019), making it imperative that the public health and entomology communities bridge this knowledge gap and study both the distribution and disease potential of Cx. stigmatosoma and Cx. thriambus.
Culex stigmatosoma was described from multiple specimens collected in California and Oregon (Dyar 1907), and Cx. thriambus from specimens collected in 1920 along the Guadalupe River near Kerrville, Kerr County, Texas (Dyar 1921). Subsequent collections provided evidence of a geographical range from the southwestern USA to northern South America for Cx. stigmatosoma and to Central America for Cx. thriambus (Knight and Stone 1977).
There has also been much taxonomic confusion related to these species. Following Dyar's work during the early years of mosquito taxonomy, multiple descriptions and names were associated with Cx. stigmatosoma and Cx. thriambus causing confusion regarding the identification and referencing of these species. After decades of changes, Eldridge and Harbach (1989, 1992) summarized the taxonomic history of the species and recommended the conservation of the specific names Cx. stigmatosoma and Cx. thriambus by the suppression of Cx. peus Speiser. In 1991 the International Commission on Zoological Nomenclature affirmed this recommendation in a published opinion (ICZN 1991). However, even with this opinion, taxonomic and identification challenges continued (Ortega-Morales et al. 2011).
During a multicounty regional surveillance study west of San Antonio, Texas, we observed multiple errors in the Texas distributional literature of these species. Some errors were minor and of no consequence, but others were the result of the confusion between Cx. stigmatosoma and Cx. thriambus taxonomy. These errors resulted in multiple errant county records, which affected the regional study, and had previously resulted in incorrect distributional information in state-level and US publications.
Upon further analysis of the Texas distributional literature regarding these species, evidence was found to correct these errant distributional records. Therefore, the aims of this study were to present that evidence and then combine the corrected records with additional records from the Texas Department of State Health Services and from larval collections made during other Texas surveillance studies.
The study was conducted by performing an extensive literature review of Texas mosquito species distributions followed by critical analysis of the data published by Eads et al. (1951), Menzies and Eads (1954), Breland (1957), and Hill et al. (1958). Curated specimens in the Osmund P. Breland Collection at The University of Texas Biodiversity Collection, Austin, Texas, were accessed to cross-check the literature review. Additional species records were found in Fournier and Snyder (1977) and Fournier et al. (1989). Unpublished records from the Texas Department of State Health Services (Texas DSHS) and from study author-made larval collections conducted in support of multiple regional surveillance studies were also utilized.
The initial source of confounded data occurred in Eads et al. (1951), when they reported Cx. stigmatosoma from 16 Texas counties and no records for Cx. thriambus. In a follow-up article, Menzies and Eads (1954) intended to correct the data error in Eads et al. Menzies and Eads explained that they thought Cx. thriambus was a synonym of Cx. stigmatosoma. When they realized this assumption was incorrect, they reviewed their specimens and determined that only 2 of the specimens from 16 Texas county records were Cx. stigmatosoma and the other 14 records were Cx. thriambus (Menzies and Eads 1954). Unfortunately, Menzies and Eads did not delineate which records should be changed. They further added to the confusion by including a distribution map of pinned adult data from their collection, which did not correlate with the slide-mounted county records data listed in Eads et al. (1951).
Based on the statement in Menzies and Eads (1954), 2 conclusions could be drawn. First, if the Menzies and Eads statement refers to individual specimens, then 2 of the slides (n = unknown no. of slides) were Cx. stigmatosoma. Second, if the statement refers to county records, not individual slides, then 14 of the 16 records should have originally been listed as Cx. thriambus.
This first conclusion would suggest that Eads et al. (1951) collected Cx. stigmatosoma in either 1 or 2 counties. To evaluate this hypothesis, an attempt was made to locate the slides related to that work at the University of Texas Biodiversity Collection. Unfortunately, the slides were not found. Due to the inability to find evidence for the first conclusion, this study subscribed to the second and sought out evidence of the errors and corrections.
Data errors appeared in 3 comprehensive county-level species lists: Hill et al. (1958), Fournier and Snyder (1977), and Fournier et al. (1989). These compilations were important to the study because they attempted to reference the source of each county record and thereby provided a document trail for review. Most notably, Hill et al. (1958) included sufficient evidence to allow a logical deduction to correct these data.
Regarding the Cx. stigmatosoma data, Hill et al. (1958) stated that Breland (1957) had compiled the first mentioned records of this species in Brewster and Jeff Davis Counties. While Breland presented detailed larval characteristics for separating Cx. stigmatosoma and Cx. quinquefasciatus Say in Jeff Davis County (pinned specimens are in the Breland Mosquito Collection), he did not mention Brewster County.
Since Brewster County was not mentioned in Breland (1957), the study authors suspected that the Brewster County Cx. stigmatosoma record in Hill et al. (1958) was provided via a personal communication with Breland based upon a pinned specimen found in the Breland Mosquito Collection. The Cx. stigmatosoma specimen was labeled as collected in Brewster County on September 26, 1944. The collector was not listed on the specimen label, but the collection date is the same as one found in Menzies and Eads (1954) and attributed to D. R. Lindsay and L. J. Ogden, who, as assistant sanitarians, were involved in mosquito surveillance for the Texas Department of Health. Based upon this evidence, the study authors attributed the Brewster County record to Lindsay and Ogden as listed in Menzies and Eads.
Additionally, Hill et al. (1958) cited Eads et al. (1951) and used all 16 Cx. stigmatosoma records and did not mention or discuss the changes to 14 of the 16 Cx. thriambus records published by Menzies and Eads (1954). Therefore, 14 of the Cx. stigmatosoma records in Hill et al. (1958) were incorrect. Hill et al. also listed 3 new Cx. stigmatosoma records from Shelby, (R. B. Eads, Records and Files), El Paso, and Terry Counties (Menzies and Eads 1954). Like the evidence for Brewster County, the study authors found a pinned Cx. stigmatosoma specimen for Terry County in the Breland Mosquito Collection to confirm this record. No collectors were listed, but the date on the specimen was the same as those attributed to D. R. Lindsay and L. J. Ogden, as reported by Menzies and Eads.
For Cx. thriambus, Hill et al. (1958) listed unpublished records from Cameron County (Texas A&M University), from Hill and Tarrant Counties (R. H. Runder), and from Blanco, Gonzales, and Jeff Davis Counties (O. P. Breland). The Blanco, Gonzales, and Jeff Davis County records appear to have been provided to Hill et al. via personal communication with Breland and are all supported by pinned specimens in the Breland Mosquito Collection. In the Breland Mosquito Collection, the earliest date (Sep. 27, 1944) for the Jeff Davis County record is from a pinned specimen. The label did not list the collectors, but the date and labeling correlate with other collections attributed to D. R. Lindsay and L. J. Ogden, and this record was attributed to them. Additional Gonzales County collections were reported in Breland (1954a, 1954b).
Hill et al. (1958) also listed unpublished Cx. thriambus military collection records for Bell, Coryell, and Travis Counties and published military records from Rueger and Druce (1950) who reported records for Bexar, Brown, Parker, and Webb Counties. The Menzies and Eads (1954) records cited by Hill et al. also included Baylor, Coleman, Jack, Kimble, McCulloch, Runnels, and Sutton Counties. A comparison between the Cx. thriambus records in Eads et al. (1951) and those in Hill et al. (1958) showed 3 counties (Jim Wells, Shackleford, and Uvalde Counties), which did not overlap between the 2 lists.
Although Hill et al. (1958) provided evidence of which 14 county records required correction, there is a discrepancy regarding the Jim Wells and Jim Hogg County data. In Eads et al. (1951), there are 17 mosquito species records, including Cx. stigmatosoma, for Jim Wells County, and those records are included in Hill et al. There are no Jim Hogg County records in Eads et al., and the only Jim Hogg County mosquito species record in a peer-reviewed publication or government document is of Aedes aegypti in Womack (1993). No pinned specimens, for either county, were in the University of Texas, Texas A&M University, or Sam Houston State University mosquito collections.
The 4 Jim Hogg County records in Hill et al. (1958) are attributed to R. B. Eads (Records and Files), which suggests a personal communication and therefore a chance for a transcription error. A potential explanation for the Jim Hogg County records in Hill et al. is the fact that Jim Hogg and Jim Wells Counties are located near each other, and their similar sounding names could have been confused by Eads and Hill et al. Since 3 of the species, Cx. quinquefasciatus, Anopheles quadrimaculatus (Say), and Cx. tarsalis Coquillett, listed in Hill et al. are common throughout the region, and the study authors have additional Jim Hogg County collection data to support these records, the study authors recommend leaving them unchanged.
However, there is no evidence to support the Cx. thriambus record for Jim Hogg County. The study authors believe the placement of Cx. thriambus in Jim Hogg County was an attempt to replace the Jim Wells County Cx. stigmatosoma record with Cx. thriambus, but the county names were switched. Therefore, the study authors attributed the Cx. thriambus record to Jim Wells County and removed it from Jim Hogg County. To check this hypothesis, W.S. made 5 mosquito collection trips to Jim Hogg County (Nov. 2017–Aug. 2020). While 20 species were collected, the author did not collect Cx. stigmatosoma or Cx. thriambus.
With the Cx. thriambus record changed from Jim Hogg to Jim Wells County, there are now 14 counties listed in Eads et al. (1951) that correlate with the Cx. thriambus listings from Menzies and Eads (1954) and R. B. Eads (Records and Files) as presented in Hill et al. (1958). The 14 Cx. thriambus county records that should have been reported in Eads et al. are Brewster, Coleman, Edwards, Jim Wells (as corrected from Jim Hogg), Kerr, Kimble, Mason, McCulloch, Pecos, Reeves, Runnels, Sutton, Tom Green, and Wheeler Counties. The 2 Cx. stigmatosoma county records that should have been reported in Eads et al. are Uvalde and Shackelford Counties. These last 2 counties were also listed in Menzies and Eads (1954).
These recommended changes to the historical records of Cx. stigmatosoma and Cx. thriambus are significant because Fournier and Snyder (1977) and Fournier et al. (1989) included these errors. These compilations also added new records. Fournier and Snyder (1977) added 2 records for Cx. stigmatosoma (Hockley, San Patricio County) and 1 record for Cx. thriambus (Kendall County). Fournier et al. (1989) added 3 records for Cx. thriambus in Fort Bend, Hidalgo, and Wichita Counties and no new records for Cx. stigmatosoma. Records in these documents were attributed to Fournier (Records and Files), which were assumed to be based upon identifications of county submissions to the Texas DSHS laboratory in Austin. These records are also included in Table 1.
In reviewing more recent literature for this study, Bradford et al. (2008) listed a Lubbock County record for Cx. thriambus based upon a misinterpretation of Sublette and Sublette (1970). Sublette and Sublette did not collect Cx. thriambus in Lubbock County, but referenced the Cx. thriambus data in Hill et al. (1958), which did not include Lubbock County. Therefore, no Cx. thriambus collection record exists for Lubbock County.
In addition to the changes outlined above, 8 Cx. stigmatosoma (21 total) and 13 Cx. thriambus (57 total) records are reported in Table 1 and represented in Fig. 1. These records were derived from county submissions to Texas DSHS (2002–2019) or study author-made collections in support of other regional surveillance studies.
The results of this study correct county-level record errors and should reduce confusion between interpreting Eads et al. (1951) and Menzies and Eads (1954). These corrections clear the path for future correct reporting of these species in state, regional, and US publications.
Before this study, Cx. stigmatosoma was last reported in Texas by Fournier and Snyder (1977), and although there are now more collection records for Cx. thriambus in Texas, little is known about their bionomics or vector potential in Texas and the southwestern USA. The updated county-level records in Table 1 establish a baseline for future studies, discussions, and publications on these species.
The authors thank Briana G. Gonzales (Jim Hogg County Texas AgriLife Extension Agent) and Erasmo Montemayor, District Conservationist, USDA Hebbronville Service Center for help in finding collecting locations in Jim Hogg County. We thank Anthony Bosworth, Nina Dacko, Whitney Qualls, and Jason Williams for providing information regarding mosquito distributions in Denton, Jim Wells, Johnson, Tarrant, and Titus Counties. The authors also thank Bill Barfield, Bryce D. Blackman, Briana Gonzales, and Steven N. Hogue Jr. for their assistance with collections in Jim Hogg County. The authors thank the participants in the Texas DSHS Arbovirus Surveillance Program for their diligence in protecting the people in their communities. The authors thank mapchart.net for allowing access and use of their online mapping capabilities in the creation of the species distribution maps.
US Army (retired), PO Box 547, Leakey, TX 78873.
Arbovirus-Entomology Team, Laboratory Services Section, Texas Department of State Health Services, 1100 W. 49th Street, Austin, TX 78756.
US Army (retired), 2911 Bandera Hwy, Kerrville, TX 78028.
Curator of Entomology, University of Texas Biodiversity Collections, 3001 Lake Austin Blvd, Brackenridge Field Lab, Austin, TX 78703.